pfeifferj/DeepSeek-V4.1-Flash-GSQ-RCO-GGUF

zetlyn/models-gguf quantisation gguf pfeifferj/DeepSeek-V4.1-Flash-GSQ-RCO-GGUF hf deepseek-ai/DeepSeek-V4.1-Flash known 2026-09-13

https://huggingface.co/pfeifferj/DeepSeek-V4.1-Flash-GSQ-RCO-GGUF

Properties

authorpfeifferj
receipt
Source
GGUF quantisations
Its words
pfeifferj
Read by
field:author
Said since
2026-09-29 09:50 UTC
Last answered
2026-10-04 18:16 UTC
Original
open at the source
What the source handed over
{
  "_id": "6aa69da6e29d415b0da7eee6",
  "author": "pfeifferj",
  "cardData": {
    "base_model": "deepseek-ai/DeepSeek-V4.1-Flash",
    "base_model_relation": "quantized",
    "library_name": "gguf",
    "license": "mit",
    "pipeline_tag": "image-text-to-text",
    "tags": [
      "gguf",
      "gsq",
      "rco",
      "quantization",
      "mixed-precision",
      "multimodal",
      "vision"
    ]
  },
  "createdAt": "2026-09-13T12:57:10.000Z",
  "downloads": 9685,
  "gated": false,
  "id": "pfeifferj/DeepSeek-V4.1-Flash-GSQ-RCO-GGUF",
  "lastModified": "2026-09-14T20:13:49.000Z",
  "library_name": "gguf",
  "likes": 12,
  "modelId": "pfeifferj/DeepSeek-V4.1-Flash-GSQ-RCO-GGUF",
  "pipeline_tag": "image-text-to-text",
  "private": false,
  "sha": "8270aceb26eb5590ea40f347256a1736999f551d",
  "siblings": [
    {
      "rfilename": ".gitattributes"
    },
    {
      "rfilename": "DeepSeek-V4.1-Flash-DSpark-draft.gguf"
    },
    {
      "rfilename": "DeepSeek-V4.1-Flash-GSQ-RCO-3.0bit.gguf"
    },
    {
      "rfilename": "DeepSeek-V4.1-Flash-mmproj-BF16.gguf"
    },
    {
      "rfilename": "LICENSE"
    },
    {
      "rfilename": "README.md"
    },
    {
      "rfilename": "SHA256SUMS"
    },
    {
      "rfilename": "eval/2026-09-13/PPL-COMPARISON.md"
    },
    {
      "rfilename": "eval/2026-09-13/README.md"
    },
    {
      "rfilename": "eval/2026-09-13/analyze_mmlu.py"
    },
    {
      "rfilename": "eval/2026-09-13/bench/NOTICE.md"
    },
    {
      "rfilename": "eval/2026-09-13/bench/evaluate_gguf.py"
    },
    {
      "rfilename": "eval/2026-09-13/bench/instruction_following_eval/LICENSE"
    },
    {
      "rfilename": "eval/2026-09-13/bench/instruction_following_eval/evaluation_lib.py"
    },
    {
      "rfilename": "eval/2026-09-13/bench/instruction_following_eval/instructions.py"
    },
    {
      "rfilename": "eval/2026-09-13/bench/instruction_following_eval/instructions_registry.py"
    },
    {
      "rfilename": "eval/2026-09-13/bench/instruction_following_eval/instructions_util.py"
    },
    {
      "rfilename": "eval/2026-09-13/bench/mini_bench.py"
    },
    {
      "rfilename": "eval/2026-09-13/bench/request-plan.json"
    },
    {
      "rfilename": "eval/2026-09-13/bench/run_diagnostics.py"
    },
    {
      "rfilename": "eval/2026-09-13/bench/run_ppl.py"
    },
    {
      "rfilename": "eval/2026-09-13/bench/summarize_diagnostics.py"
    },
    {
      "rfilename": "eval/2026-09-13/compare_ppl_caches.py"
    },
    {
      "rfilename": "eval/2026-09-13/data/calibration-manifest.json"
    },
    {
      "rfilename": "eval/2026-09-13/data/download-manifest.json"
    },
    {
      "rfilename": "eval/2026-09-13/data/gsm8k.json"
    },
    {
      "rfilename": "eval/2026-09-13/data/heldout.txt"
    },
    {
      "rfilename": "eval/2026-09-13/data/ifeval.jsonl"
    },
    {
      "rfilename": "eval/2026-09-13/diagnostics/README.md"
    },
    {
      "rfilename": "eval/2026-09-13/diagnostics/check_scorer.py"
    },
    {
      "rfilename": "eval/2026-09-13/diagnostics/mini_bench_glm_adapter.py"
    },
    {
      "rfilename": "eval/2026-09-13/diagnostics/run_diagnostics.py"
    },
    {
      "rfilename": "eval/2026-09-13/diagnostics/summarize_diagnostics.py"
    },
    {
      "rfilename": "eval/2026-09-13/provenance/deepseek-mmlu-runtime.json"
    },
    {
      "rfilename": "eval/2026-09-13/reference/build_tasks.py"
    },
    {
      "rfilename": "eval/2026-09-13/reference/mc-score.cpp"
    },
    {
      "rfilename": "eval/2026-09-13/reference/questions.json"
    },
    {
      "rfilename": "eval/2026-09-13/reference/subset.json"
    },
    {
      "rfilename": "eval/2026-09-13/reference/task-index.json"
    },
    {
      "rfilename": "eval/2026-09-13/reference/tasks.bin"
    },
    {
      "rfilename": "eval/2026-09-13/reference/test.parquet"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-f32-run-manifest.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-f32-throughput-unvalidated.tsv"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-gpu01-parity.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-heldout-tokenizer-parity.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-middle-complete.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-native-cache-format-probe.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-oracle-few-vs-ref-f32kv.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-oracle-few-vs-ref-nofa.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-oracle-long-f32kv-boundary.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-quant-topology-2gpu-vs-4gpu.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-reference-f32-batching.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-reference-f32-short-comparison.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-reference-f32-short-nofa-comparison.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-shard-merge.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-shard-plan.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-tail-complete.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-topk-validation.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-vocab-parity.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek3.0-f32.gpu01-parity.tsv"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek3.0-f32.gpu0123-parity.tsv"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek3.0-f32.initial.tsv"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek3.0-f32.middle.tsv"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek3.0-f32.primary.tsv"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek3.0-f32.tail.tsv"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek3.0-f32.tsv"
    },
    {
      "rfilename": "eval/2026-09-13/results/diagnostics/deepseek3.0-f32.generation-probes.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/diagnostics/deepseek3.0-f32.metadata.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/diagnostics/deepseek3.0-f32.mini.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/diagnostics/deepseek3.0-f32.mini.native-specials.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/diagnostics/deepseek3.0-f32.ppl.log"
    },
    {
      "rfilename": "eval/2026-09-13/results/diagnostics/deepseek3.0-f32.ppl.validation.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/diagnostics/deepseek3.0-f32.request-plan.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/diagnostics/deepseek3.0-f32.serial-parity.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/diagnostics/deepseek3.0-f32.serial-parity.log"
    },
    {
      "rfilename": "eval/2026-09-13/results/diagnostics/deepseek3.0-f32.server0.log"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek-analysis-config.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek-analysis.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek-analysis.md"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek-f32-run-manifest.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek-f32-throughput-unvalidated.tsv"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek-gpu01-parity.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek-middle-complete.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek-shard-merge.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek-shard-plan.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek-tail-complete.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek3.0-f32.gpu01-parity.tsv"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek3.0-f32.initial.tsv"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek3.0-f32.middle.tsv"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek3.0-f32.primary.tsv"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek3.0-f32.tail.tsv"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek3.0-f32.tsv"
    },
    {
      "rfilename": "eval/2026-09-13/test_analyze_mmlu.py"
    },
    {
      "rfilename": "eval/2026-09-13/test_compare_ppl_caches.py"
    },
    {
      "rfilename": "runtime/LICENSE.DeepSeek"
    },
    {
      "rfilename": "runtime/LICENSE.llama.cpp"
    },
    {
      "rfilename": "runtime/README.md"
    },
    {
      "rfilename": "runtime/REPRODUCE.md"
    },
    {
      "rfilename": "runtime/build_helpers.sh"
    },
    {
      "rfilename": "runtime/capture.cpp"
    },
    {
      "rfilename": "runtime/deepseek-components.log"
    },
    {
      "rfilename": "runtime/deepseek-f32-throughput.log"
    },
    {
      "rfilename": "runtime/deepseek-gpu01-parity.json"
    },
    {
      "rfilename": "runtime/deepseek-heldout-tokenizer-parity.json"
    },
    {
      "rfilename": "runtime/deepseek-helper-build-check.log"
    },
    {
      "rfilename": "runtime/deepseek-native-vocab.bin"
    },
    {
      "rfilename": "runtime/deepseek-perplexity-sidecar.patch"
    },
    {
      "rfilename": "runtime/deepseek-preflight.log"
    },
    {
      "rfilename": "runtime/deepseek-quant-topology-2gpu-vs-4gpu.json"
    },
    {
      "rfilename": "runtime/deepseek-runtime-base-source.tar.gz"
    },
    {
      "rfilename": "runtime/deepseek-runtime.patch"
    },
    {
      "rfilename": "runtime/deepseek-shard-merge.json"
    },
    {
      "rfilename": "runtime/deepseek-topk-validation.json"
    },
    {
      "rfilename": "runtime/deepseek-vocab-parity.json"
    },
    {
      "rfilename": "runtime/deepseek3.0-f32.gpu0123-parity.tsv"
    },
    {
      "rfilename": "runtime/deepseek3.0-f32.tsv"
    },
    {
      "rfilename": "runtime/diagnostics/deepseek3.0-f32.generation-probes.json"
    },
    {
      "rfilename": "runtime/diagnostics/deepseek3.0-f32.metadata.json"
    },
    {
      "rfilename": "runtime/diagnostics/deepseek3.0-f32.mini.json"
    },
    {
      "rfilename": "runtime/diagnostics/deepseek3.0-f32.mini.native-specials.json"
    },
    {
      "rfilename": "runtime/diagnostics/deepseek3.0-f32.ppl.log"
    },
    {
      "rfilename": "runtime/diagnostics/deepseek3.0-f32.ppl.validation.json"
    },
    {
      "rfilename": "runtime/diagnostics/deepseek3.0-f32.request-plan.json"
    },
    {
      "rfilename": "runtime/diagnostics/deepseek3.0-f32.serial-parity.json"
    },
    {
      "rfilename": "runtime/diagnostics/deepseek3.0-f32.serial-parity.log"
    },
    {
      "rfilename": "runtime/diagnostics/deepseek3.0-f32.server0.log"
    },
    {
      "rfilename": "runtime/dump_vocab.cpp"
    },
    {
      "rfilename": "runtime/estimate_mmlu_time.py"
    },
    {
      "rfilename": "runtime/helper-build-validation.json"
    },
    {
      "rfilename": "runtime/mc-score-deepseek-f32.cpp"
    },
    {
      "rfilename": "runtime/nltk-resources.json"
    },
    {
      "rfilename": "runtime/nltk_data/tokenizers/punkt/README"
    },
    {
      "rfilename": "runtime/nltk_data/tokenizers/punkt/english.pickle"
    },
    {
      "rfilename": "runtime/nltk_data/tokenizers/punkt_tab/README"
    },
    {
      "rfilename": "runtime/nltk_data/tokenizers/punkt_tab/english/abbrev_types.txt"
    },
    {
      "rfilename": "runtime/nltk_data/tokenizers/punkt_tab/english/collocations.tab"
    },
    {
      "rfilename": "runtime/nltk_data/tokenizers/punkt_tab/english/ortho_context.tab"
    },
    {
      "rfilename": "runtime/nltk_data/tokenizers/punkt_tab/english/sent_starters.txt"
    },
    {
      "rfilename": "runtime/official-source/LICENSE"
    },
    {
      "rfilename": "runtime/official-source/config.json"
    },
    {
      "rfilename": "runtime/official-source/encoding/encoding.py"
    },
    {
      "rfilename": "runtime/official-source/inference/config.json"
    },
    {
      "rfilename": "runtime/official-source/inference/engram.py"
    },
    {
      "rfilename": "runtime/official-source/inference/image_processor.py"
    },
    {
      "rfilename": "runtime/official-source/inference/kernel.py"
    },
    {
      "rfilename": "runtime/official-source/inference/model.py"
    },
    {
      "rfilename": "runtime/official-source/inference/vision.py"
    },
    {
      "rfilename": "runtime/official-source/manifest.json"
    },
    {
      "rfilename": "runtime/official-source/tokenizer.json"
    },
    {
      "rfilename": "runtime/oracle/README.md"
    },
    {
      "rfilename": "runtime/oracle/cache_arithmetic.py"
    },
    {
      "rfilename": "runtime/oracle/compare_logits.py"
    },
    {
      "rfilename": "runtime/oracle/gguf_oracle.py"
    },
    {
      "rfilename": "runtime/oracle/results/deepseek-capture-few-ref-f32kv/logits.f32"
    },
    {
      "rfilename": "runtime/oracle/results/deepseek-capture-few-ref-f32kv/tokens.json"
    },
    {
      "rfilename": "runtime/oracle/results/deepseek-native-cache-format-probe.json"
    },
    {
      "rfilename": "runtime/oracle/results/deepseek-oracle-few-f32kv-cascade/logits.f32"
    },
    {
      "rfilename": "runtime/oracle/results/deepseek-oracle-few-f32kv-cascade/report.json"
    },
    {
      "rfilename": "runtime/oracle/results/deepseek-oracle-few-vs-ref-f32kv.json"
    },
    {
      "rfilename": "runtime/oracle/results/deepseek-oracle-long-f32kv-boundary.json"
    },
    {
      "rfilename": "runtime/oracle/results/deepseek-oracle-long-ref-f32kv-attention/report.json"
    },
    {
      "rfilename": "runtime/oracle/results/few-attention-f32-f16kv.json"
    },
    {
      "rfilename": "runtime/oracle/results/long-attention-f32-f16kv.json"
    },
    {
      "rfilename": "runtime/oracle/results/short-components-default.json"
    },
    {
      "rfilename": "runtime/oracle/results/short-components-f32.json"
    },
    {
      "rfilename": "runtime/perplexity.cpp"
    },
    {
      "rfilename": "runtime/portable-layout-validation.json"
    },
    {
      "rfilename": "runtime/prepare_replay_workspace.py"
    },
    {
      "rfilename": "runtime/provenance.json"
    },
    {
      "rfilename": "runtime/python-environment.json"
    },
    {
      "rfilename": "runtime/reference/deepseek-metadata.json"
    },
    {
      "rfilename": "runtime/reference/deepseek41.cpp"
    },
    {
      "rfilename": "runtime/reference/encoding-files.json"
    },
    {
      "rfilename": "runtime/reference/engram.py"
    },
    {
      "rfilename": "runtime/reference/model.py"
    },
    {
      "rfilename": "runtime/reference/subset.json"
    },
    {
      "rfilename": "runtime/reference/task-index.json"
    },
    {
      "rfilename": "runtime/reference/tasks.bin"
    },
    {
      "rfilename": "runtime/requirements-environment.txt"
    },
    {
      "rfilename": "runtime/requirements-minimal.txt"
    },
    {
      "rfilename": "runtime/rescore_diagnostics.py"
    },
    {
      "rfilename": "runtime/research/README.md"
    },
    {
      "rfilename": "runtime/research/deployment-crosscheck.md"
    },
    {
      "rfilename": "runtime/research/engine-implementations.md"
    },
    {
      "rfilename": "runtime/research/jigsaw_report.json"
    },
    {
      "rfilename": "runtime/research/jigsaw_runtime.json"
    },
    {
      "rfilename": "runtime/research/llama_conversion_pr.json"
    },
    {
      "rfilename": "runtime/research/miaai_runtime.json"
    },
    {
      "rfilename": "runtime/research/official-runtime.md"
    },
    {
      "rfilename": "runtime/research/question-state-isolation.md"
    },
    {
      "rfilename": "runtime/research/sglang_runtime_pr.json"
    },
    {
      "rfilename": "runtime/research/shi3z_runtime.json"
    },
    {
      "rfilename": "runtime/research/sources/DeepSeek_V41_Tech_Report.pdf"
    },
    {
      "rfilename": "runtime/research/sources/DeepSeek_V41_Tech_Report.txt"
    },
    {
      "rfilename": "runtime/research/sources/LICENSE"
    },
    {
      "rfilename": "runtime/research/sources/README.md"
    },
    {
      "rfilename": "runtime/research/sources/encoding/README.md"
    },
    {
      "rfilename": "runtime/research/sources/encoding_README.md"
    },
    {
      "rfilename": "runtime/research/sources/evaluation/README.md"
    },
    {
      "rfilename": "runtime/research/sources/evaluation_README.md"
    },
    {
      "rfilename": "runtime/research/sources/inference/README.md"
    },
    {
      "rfilename": "runtime/research/sources/inference_README.md"
    },
    {
      "rfilename": "runtime/research/sources/inference_config.json"
    },
    {
      "rfilename": "runtime/research/sources/inference_engram.py"
    },
    {
      "rfilename": "runtime/research/sources/inference_kernel.py"
    },
    {
      "rfilename": "runtime/research/sources/inference_model.py"
    },
    {
      "rfilename": "runtime/research/sources/manifest.json"
    },
    {
      "rfilename": "runtime/research/vcruz305_runtime.json"
    },
    {
      "rfilename": "runtime/run_diagnostics.executed.py"
    },
    {
      "rfilename": "runtime/run_diagnostics.py"
    },
    {
      "rfilename": "runtime/run_sharded_mmlu.py"
    },
    {
      "rfilename": "runtime/run_topology_gate.py"
    },
    {
      "rfilename": "runtime/test_reference_components.py"
    },
    {
      "rfilename": "runtime/validate_ppl.py"
    },
    {
      "rfilename": "runtime/validate_topk_masks.py"
    },
    {
      "rfilename": "runtime/verify_vocab.py"
    }
  ],
  "tags": [
    "gguf",
    "gsq",
    "rco",
    "quantization",
    "mixed-precision",
    "multimodal",
    "vision",
    "image-text-to-text",
    "arxiv:2604.18556",
    "arxiv:2605.00649",
    "base_model:deepseek-ai/DeepSeek-V4.1-Flash",
    "base_model:quantized:deepseek-ai/DeepSeek-V4.1-Flash",
    "doi:10.57967/hf/10402",
    "license:mit",
    "endpoints_compatible",
    "region:us",
    "conversational"
  ]
}
basedeepseek-ai/DeepSeek-V4.1-Flash
receipt
Source
GGUF quantisations
Its words
deepseek-ai/DeepSeek-V4.1-Flash
Read by
field:cardData.base_model[]
Said since
2026-09-29 09:50 UTC
Last answered
2026-10-04 18:16 UTC
Original
open at the source
What the source handed over
{
  "_id": "6aa69da6e29d415b0da7eee6",
  "author": "pfeifferj",
  "cardData": {
    "base_model": "deepseek-ai/DeepSeek-V4.1-Flash",
    "base_model_relation": "quantized",
    "library_name": "gguf",
    "license": "mit",
    "pipeline_tag": "image-text-to-text",
    "tags": [
      "gguf",
      "gsq",
      "rco",
      "quantization",
      "mixed-precision",
      "multimodal",
      "vision"
    ]
  },
  "createdAt": "2026-09-13T12:57:10.000Z",
  "downloads": 9685,
  "gated": false,
  "id": "pfeifferj/DeepSeek-V4.1-Flash-GSQ-RCO-GGUF",
  "lastModified": "2026-09-14T20:13:49.000Z",
  "library_name": "gguf",
  "likes": 12,
  "modelId": "pfeifferj/DeepSeek-V4.1-Flash-GSQ-RCO-GGUF",
  "pipeline_tag": "image-text-to-text",
  "private": false,
  "sha": "8270aceb26eb5590ea40f347256a1736999f551d",
  "siblings": [
    {
      "rfilename": ".gitattributes"
    },
    {
      "rfilename": "DeepSeek-V4.1-Flash-DSpark-draft.gguf"
    },
    {
      "rfilename": "DeepSeek-V4.1-Flash-GSQ-RCO-3.0bit.gguf"
    },
    {
      "rfilename": "DeepSeek-V4.1-Flash-mmproj-BF16.gguf"
    },
    {
      "rfilename": "LICENSE"
    },
    {
      "rfilename": "README.md"
    },
    {
      "rfilename": "SHA256SUMS"
    },
    {
      "rfilename": "eval/2026-09-13/PPL-COMPARISON.md"
    },
    {
      "rfilename": "eval/2026-09-13/README.md"
    },
    {
      "rfilename": "eval/2026-09-13/analyze_mmlu.py"
    },
    {
      "rfilename": "eval/2026-09-13/bench/NOTICE.md"
    },
    {
      "rfilename": "eval/2026-09-13/bench/evaluate_gguf.py"
    },
    {
      "rfilename": "eval/2026-09-13/bench/instruction_following_eval/LICENSE"
    },
    {
      "rfilename": "eval/2026-09-13/bench/instruction_following_eval/evaluation_lib.py"
    },
    {
      "rfilename": "eval/2026-09-13/bench/instruction_following_eval/instructions.py"
    },
    {
      "rfilename": "eval/2026-09-13/bench/instruction_following_eval/instructions_registry.py"
    },
    {
      "rfilename": "eval/2026-09-13/bench/instruction_following_eval/instructions_util.py"
    },
    {
      "rfilename": "eval/2026-09-13/bench/mini_bench.py"
    },
    {
      "rfilename": "eval/2026-09-13/bench/request-plan.json"
    },
    {
      "rfilename": "eval/2026-09-13/bench/run_diagnostics.py"
    },
    {
      "rfilename": "eval/2026-09-13/bench/run_ppl.py"
    },
    {
      "rfilename": "eval/2026-09-13/bench/summarize_diagnostics.py"
    },
    {
      "rfilename": "eval/2026-09-13/compare_ppl_caches.py"
    },
    {
      "rfilename": "eval/2026-09-13/data/calibration-manifest.json"
    },
    {
      "rfilename": "eval/2026-09-13/data/download-manifest.json"
    },
    {
      "rfilename": "eval/2026-09-13/data/gsm8k.json"
    },
    {
      "rfilename": "eval/2026-09-13/data/heldout.txt"
    },
    {
      "rfilename": "eval/2026-09-13/data/ifeval.jsonl"
    },
    {
      "rfilename": "eval/2026-09-13/diagnostics/README.md"
    },
    {
      "rfilename": "eval/2026-09-13/diagnostics/check_scorer.py"
    },
    {
      "rfilename": "eval/2026-09-13/diagnostics/mini_bench_glm_adapter.py"
    },
    {
      "rfilename": "eval/2026-09-13/diagnostics/run_diagnostics.py"
    },
    {
      "rfilename": "eval/2026-09-13/diagnostics/summarize_diagnostics.py"
    },
    {
      "rfilename": "eval/2026-09-13/provenance/deepseek-mmlu-runtime.json"
    },
    {
      "rfilename": "eval/2026-09-13/reference/build_tasks.py"
    },
    {
      "rfilename": "eval/2026-09-13/reference/mc-score.cpp"
    },
    {
      "rfilename": "eval/2026-09-13/reference/questions.json"
    },
    {
      "rfilename": "eval/2026-09-13/reference/subset.json"
    },
    {
      "rfilename": "eval/2026-09-13/reference/task-index.json"
    },
    {
      "rfilename": "eval/2026-09-13/reference/tasks.bin"
    },
    {
      "rfilename": "eval/2026-09-13/reference/test.parquet"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-f32-run-manifest.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-f32-throughput-unvalidated.tsv"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-gpu01-parity.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-heldout-tokenizer-parity.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-middle-complete.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-native-cache-format-probe.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-oracle-few-vs-ref-f32kv.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-oracle-few-vs-ref-nofa.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-oracle-long-f32kv-boundary.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-quant-topology-2gpu-vs-4gpu.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-reference-f32-batching.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-reference-f32-short-comparison.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-reference-f32-short-nofa-comparison.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-shard-merge.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-shard-plan.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-tail-complete.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-topk-validation.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-vocab-parity.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek3.0-f32.gpu01-parity.tsv"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek3.0-f32.gpu0123-parity.tsv"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek3.0-f32.initial.tsv"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek3.0-f32.middle.tsv"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek3.0-f32.primary.tsv"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek3.0-f32.tail.tsv"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek3.0-f32.tsv"
    },
    {
      "rfilename": "eval/2026-09-13/results/diagnostics/deepseek3.0-f32.generation-probes.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/diagnostics/deepseek3.0-f32.metadata.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/diagnostics/deepseek3.0-f32.mini.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/diagnostics/deepseek3.0-f32.mini.native-specials.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/diagnostics/deepseek3.0-f32.ppl.log"
    },
    {
      "rfilename": "eval/2026-09-13/results/diagnostics/deepseek3.0-f32.ppl.validation.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/diagnostics/deepseek3.0-f32.request-plan.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/diagnostics/deepseek3.0-f32.serial-parity.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/diagnostics/deepseek3.0-f32.serial-parity.log"
    },
    {
      "rfilename": "eval/2026-09-13/results/diagnostics/deepseek3.0-f32.server0.log"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek-analysis-config.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek-analysis.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek-analysis.md"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek-f32-run-manifest.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek-f32-throughput-unvalidated.tsv"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek-gpu01-parity.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek-middle-complete.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek-shard-merge.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek-shard-plan.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek-tail-complete.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek3.0-f32.gpu01-parity.tsv"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek3.0-f32.initial.tsv"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek3.0-f32.middle.tsv"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek3.0-f32.primary.tsv"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek3.0-f32.tail.tsv"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek3.0-f32.tsv"
    },
    {
      "rfilename": "eval/2026-09-13/test_analyze_mmlu.py"
    },
    {
      "rfilename": "eval/2026-09-13/test_compare_ppl_caches.py"
    },
    {
      "rfilename": "runtime/LICENSE.DeepSeek"
    },
    {
      "rfilename": "runtime/LICENSE.llama.cpp"
    },
    {
      "rfilename": "runtime/README.md"
    },
    {
      "rfilename": "runtime/REPRODUCE.md"
    },
    {
      "rfilename": "runtime/build_helpers.sh"
    },
    {
      "rfilename": "runtime/capture.cpp"
    },
    {
      "rfilename": "runtime/deepseek-components.log"
    },
    {
      "rfilename": "runtime/deepseek-f32-throughput.log"
    },
    {
      "rfilename": "runtime/deepseek-gpu01-parity.json"
    },
    {
      "rfilename": "runtime/deepseek-heldout-tokenizer-parity.json"
    },
    {
      "rfilename": "runtime/deepseek-helper-build-check.log"
    },
    {
      "rfilename": "runtime/deepseek-native-vocab.bin"
    },
    {
      "rfilename": "runtime/deepseek-perplexity-sidecar.patch"
    },
    {
      "rfilename": "runtime/deepseek-preflight.log"
    },
    {
      "rfilename": "runtime/deepseek-quant-topology-2gpu-vs-4gpu.json"
    },
    {
      "rfilename": "runtime/deepseek-runtime-base-source.tar.gz"
    },
    {
      "rfilename": "runtime/deepseek-runtime.patch"
    },
    {
      "rfilename": "runtime/deepseek-shard-merge.json"
    },
    {
      "rfilename": "runtime/deepseek-topk-validation.json"
    },
    {
      "rfilename": "runtime/deepseek-vocab-parity.json"
    },
    {
      "rfilename": "runtime/deepseek3.0-f32.gpu0123-parity.tsv"
    },
    {
      "rfilename": "runtime/deepseek3.0-f32.tsv"
    },
    {
      "rfilename": "runtime/diagnostics/deepseek3.0-f32.generation-probes.json"
    },
    {
      "rfilename": "runtime/diagnostics/deepseek3.0-f32.metadata.json"
    },
    {
      "rfilename": "runtime/diagnostics/deepseek3.0-f32.mini.json"
    },
    {
      "rfilename": "runtime/diagnostics/deepseek3.0-f32.mini.native-specials.json"
    },
    {
      "rfilename": "runtime/diagnostics/deepseek3.0-f32.ppl.log"
    },
    {
      "rfilename": "runtime/diagnostics/deepseek3.0-f32.ppl.validation.json"
    },
    {
      "rfilename": "runtime/diagnostics/deepseek3.0-f32.request-plan.json"
    },
    {
      "rfilename": "runtime/diagnostics/deepseek3.0-f32.serial-parity.json"
    },
    {
      "rfilename": "runtime/diagnostics/deepseek3.0-f32.serial-parity.log"
    },
    {
      "rfilename": "runtime/diagnostics/deepseek3.0-f32.server0.log"
    },
    {
      "rfilename": "runtime/dump_vocab.cpp"
    },
    {
      "rfilename": "runtime/estimate_mmlu_time.py"
    },
    {
      "rfilename": "runtime/helper-build-validation.json"
    },
    {
      "rfilename": "runtime/mc-score-deepseek-f32.cpp"
    },
    {
      "rfilename": "runtime/nltk-resources.json"
    },
    {
      "rfilename": "runtime/nltk_data/tokenizers/punkt/README"
    },
    {
      "rfilename": "runtime/nltk_data/tokenizers/punkt/english.pickle"
    },
    {
      "rfilename": "runtime/nltk_data/tokenizers/punkt_tab/README"
    },
    {
      "rfilename": "runtime/nltk_data/tokenizers/punkt_tab/english/abbrev_types.txt"
    },
    {
      "rfilename": "runtime/nltk_data/tokenizers/punkt_tab/english/collocations.tab"
    },
    {
      "rfilename": "runtime/nltk_data/tokenizers/punkt_tab/english/ortho_context.tab"
    },
    {
      "rfilename": "runtime/nltk_data/tokenizers/punkt_tab/english/sent_starters.txt"
    },
    {
      "rfilename": "runtime/official-source/LICENSE"
    },
    {
      "rfilename": "runtime/official-source/config.json"
    },
    {
      "rfilename": "runtime/official-source/encoding/encoding.py"
    },
    {
      "rfilename": "runtime/official-source/inference/config.json"
    },
    {
      "rfilename": "runtime/official-source/inference/engram.py"
    },
    {
      "rfilename": "runtime/official-source/inference/image_processor.py"
    },
    {
      "rfilename": "runtime/official-source/inference/kernel.py"
    },
    {
      "rfilename": "runtime/official-source/inference/model.py"
    },
    {
      "rfilename": "runtime/official-source/inference/vision.py"
    },
    {
      "rfilename": "runtime/official-source/manifest.json"
    },
    {
      "rfilename": "runtime/official-source/tokenizer.json"
    },
    {
      "rfilename": "runtime/oracle/README.md"
    },
    {
      "rfilename": "runtime/oracle/cache_arithmetic.py"
    },
    {
      "rfilename": "runtime/oracle/compare_logits.py"
    },
    {
      "rfilename": "runtime/oracle/gguf_oracle.py"
    },
    {
      "rfilename": "runtime/oracle/results/deepseek-capture-few-ref-f32kv/logits.f32"
    },
    {
      "rfilename": "runtime/oracle/results/deepseek-capture-few-ref-f32kv/tokens.json"
    },
    {
      "rfilename": "runtime/oracle/results/deepseek-native-cache-format-probe.json"
    },
    {
      "rfilename": "runtime/oracle/results/deepseek-oracle-few-f32kv-cascade/logits.f32"
    },
    {
      "rfilename": "runtime/oracle/results/deepseek-oracle-few-f32kv-cascade/report.json"
    },
    {
      "rfilename": "runtime/oracle/results/deepseek-oracle-few-vs-ref-f32kv.json"
    },
    {
      "rfilename": "runtime/oracle/results/deepseek-oracle-long-f32kv-boundary.json"
    },
    {
      "rfilename": "runtime/oracle/results/deepseek-oracle-long-ref-f32kv-attention/report.json"
    },
    {
      "rfilename": "runtime/oracle/results/few-attention-f32-f16kv.json"
    },
    {
      "rfilename": "runtime/oracle/results/long-attention-f32-f16kv.json"
    },
    {
      "rfilename": "runtime/oracle/results/short-components-default.json"
    },
    {
      "rfilename": "runtime/oracle/results/short-components-f32.json"
    },
    {
      "rfilename": "runtime/perplexity.cpp"
    },
    {
      "rfilename": "runtime/portable-layout-validation.json"
    },
    {
      "rfilename": "runtime/prepare_replay_workspace.py"
    },
    {
      "rfilename": "runtime/provenance.json"
    },
    {
      "rfilename": "runtime/python-environment.json"
    },
    {
      "rfilename": "runtime/reference/deepseek-metadata.json"
    },
    {
      "rfilename": "runtime/reference/deepseek41.cpp"
    },
    {
      "rfilename": "runtime/reference/encoding-files.json"
    },
    {
      "rfilename": "runtime/reference/engram.py"
    },
    {
      "rfilename": "runtime/reference/model.py"
    },
    {
      "rfilename": "runtime/reference/subset.json"
    },
    {
      "rfilename": "runtime/reference/task-index.json"
    },
    {
      "rfilename": "runtime/reference/tasks.bin"
    },
    {
      "rfilename": "runtime/requirements-environment.txt"
    },
    {
      "rfilename": "runtime/requirements-minimal.txt"
    },
    {
      "rfilename": "runtime/rescore_diagnostics.py"
    },
    {
      "rfilename": "runtime/research/README.md"
    },
    {
      "rfilename": "runtime/research/deployment-crosscheck.md"
    },
    {
      "rfilename": "runtime/research/engine-implementations.md"
    },
    {
      "rfilename": "runtime/research/jigsaw_report.json"
    },
    {
      "rfilename": "runtime/research/jigsaw_runtime.json"
    },
    {
      "rfilename": "runtime/research/llama_conversion_pr.json"
    },
    {
      "rfilename": "runtime/research/miaai_runtime.json"
    },
    {
      "rfilename": "runtime/research/official-runtime.md"
    },
    {
      "rfilename": "runtime/research/question-state-isolation.md"
    },
    {
      "rfilename": "runtime/research/sglang_runtime_pr.json"
    },
    {
      "rfilename": "runtime/research/shi3z_runtime.json"
    },
    {
      "rfilename": "runtime/research/sources/DeepSeek_V41_Tech_Report.pdf"
    },
    {
      "rfilename": "runtime/research/sources/DeepSeek_V41_Tech_Report.txt"
    },
    {
      "rfilename": "runtime/research/sources/LICENSE"
    },
    {
      "rfilename": "runtime/research/sources/README.md"
    },
    {
      "rfilename": "runtime/research/sources/encoding/README.md"
    },
    {
      "rfilename": "runtime/research/sources/encoding_README.md"
    },
    {
      "rfilename": "runtime/research/sources/evaluation/README.md"
    },
    {
      "rfilename": "runtime/research/sources/evaluation_README.md"
    },
    {
      "rfilename": "runtime/research/sources/inference/README.md"
    },
    {
      "rfilename": "runtime/research/sources/inference_README.md"
    },
    {
      "rfilename": "runtime/research/sources/inference_config.json"
    },
    {
      "rfilename": "runtime/research/sources/inference_engram.py"
    },
    {
      "rfilename": "runtime/research/sources/inference_kernel.py"
    },
    {
      "rfilename": "runtime/research/sources/inference_model.py"
    },
    {
      "rfilename": "runtime/research/sources/manifest.json"
    },
    {
      "rfilename": "runtime/research/vcruz305_runtime.json"
    },
    {
      "rfilename": "runtime/run_diagnostics.executed.py"
    },
    {
      "rfilename": "runtime/run_diagnostics.py"
    },
    {
      "rfilename": "runtime/run_sharded_mmlu.py"
    },
    {
      "rfilename": "runtime/run_topology_gate.py"
    },
    {
      "rfilename": "runtime/test_reference_components.py"
    },
    {
      "rfilename": "runtime/validate_ppl.py"
    },
    {
      "rfilename": "runtime/validate_topk_masks.py"
    },
    {
      "rfilename": "runtime/verify_vocab.py"
    }
  ],
  "tags": [
    "gguf",
    "gsq",
    "rco",
    "quantization",
    "mixed-precision",
    "multimodal",
    "vision",
    "image-text-to-text",
    "arxiv:2604.18556",
    "arxiv:2605.00649",
    "base_model:deepseek-ai/DeepSeek-V4.1-Flash",
    "base_model:quantized:deepseek-ai/DeepSeek-V4.1-Flash",
    "doi:10.57967/hf/10402",
    "license:mit",
    "endpoints_compatible",
    "region:us",
    "conversational"
  ]
}
downloads9685
receipt
Source
GGUF quantisations
Its words
9685
Read by
field:downloads
Said since
2026-09-29 09:50 UTC
Last answered
2026-10-04 18:16 UTC
Original
open at the source
What the source handed over
{
  "_id": "6aa69da6e29d415b0da7eee6",
  "author": "pfeifferj",
  "cardData": {
    "base_model": "deepseek-ai/DeepSeek-V4.1-Flash",
    "base_model_relation": "quantized",
    "library_name": "gguf",
    "license": "mit",
    "pipeline_tag": "image-text-to-text",
    "tags": [
      "gguf",
      "gsq",
      "rco",
      "quantization",
      "mixed-precision",
      "multimodal",
      "vision"
    ]
  },
  "createdAt": "2026-09-13T12:57:10.000Z",
  "downloads": 9685,
  "gated": false,
  "id": "pfeifferj/DeepSeek-V4.1-Flash-GSQ-RCO-GGUF",
  "lastModified": "2026-09-14T20:13:49.000Z",
  "library_name": "gguf",
  "likes": 12,
  "modelId": "pfeifferj/DeepSeek-V4.1-Flash-GSQ-RCO-GGUF",
  "pipeline_tag": "image-text-to-text",
  "private": false,
  "sha": "8270aceb26eb5590ea40f347256a1736999f551d",
  "siblings": [
    {
      "rfilename": ".gitattributes"
    },
    {
      "rfilename": "DeepSeek-V4.1-Flash-DSpark-draft.gguf"
    },
    {
      "rfilename": "DeepSeek-V4.1-Flash-GSQ-RCO-3.0bit.gguf"
    },
    {
      "rfilename": "DeepSeek-V4.1-Flash-mmproj-BF16.gguf"
    },
    {
      "rfilename": "LICENSE"
    },
    {
      "rfilename": "README.md"
    },
    {
      "rfilename": "SHA256SUMS"
    },
    {
      "rfilename": "eval/2026-09-13/PPL-COMPARISON.md"
    },
    {
      "rfilename": "eval/2026-09-13/README.md"
    },
    {
      "rfilename": "eval/2026-09-13/analyze_mmlu.py"
    },
    {
      "rfilename": "eval/2026-09-13/bench/NOTICE.md"
    },
    {
      "rfilename": "eval/2026-09-13/bench/evaluate_gguf.py"
    },
    {
      "rfilename": "eval/2026-09-13/bench/instruction_following_eval/LICENSE"
    },
    {
      "rfilename": "eval/2026-09-13/bench/instruction_following_eval/evaluation_lib.py"
    },
    {
      "rfilename": "eval/2026-09-13/bench/instruction_following_eval/instructions.py"
    },
    {
      "rfilename": "eval/2026-09-13/bench/instruction_following_eval/instructions_registry.py"
    },
    {
      "rfilename": "eval/2026-09-13/bench/instruction_following_eval/instructions_util.py"
    },
    {
      "rfilename": "eval/2026-09-13/bench/mini_bench.py"
    },
    {
      "rfilename": "eval/2026-09-13/bench/request-plan.json"
    },
    {
      "rfilename": "eval/2026-09-13/bench/run_diagnostics.py"
    },
    {
      "rfilename": "eval/2026-09-13/bench/run_ppl.py"
    },
    {
      "rfilename": "eval/2026-09-13/bench/summarize_diagnostics.py"
    },
    {
      "rfilename": "eval/2026-09-13/compare_ppl_caches.py"
    },
    {
      "rfilename": "eval/2026-09-13/data/calibration-manifest.json"
    },
    {
      "rfilename": "eval/2026-09-13/data/download-manifest.json"
    },
    {
      "rfilename": "eval/2026-09-13/data/gsm8k.json"
    },
    {
      "rfilename": "eval/2026-09-13/data/heldout.txt"
    },
    {
      "rfilename": "eval/2026-09-13/data/ifeval.jsonl"
    },
    {
      "rfilename": "eval/2026-09-13/diagnostics/README.md"
    },
    {
      "rfilename": "eval/2026-09-13/diagnostics/check_scorer.py"
    },
    {
      "rfilename": "eval/2026-09-13/diagnostics/mini_bench_glm_adapter.py"
    },
    {
      "rfilename": "eval/2026-09-13/diagnostics/run_diagnostics.py"
    },
    {
      "rfilename": "eval/2026-09-13/diagnostics/summarize_diagnostics.py"
    },
    {
      "rfilename": "eval/2026-09-13/provenance/deepseek-mmlu-runtime.json"
    },
    {
      "rfilename": "eval/2026-09-13/reference/build_tasks.py"
    },
    {
      "rfilename": "eval/2026-09-13/reference/mc-score.cpp"
    },
    {
      "rfilename": "eval/2026-09-13/reference/questions.json"
    },
    {
      "rfilename": "eval/2026-09-13/reference/subset.json"
    },
    {
      "rfilename": "eval/2026-09-13/reference/task-index.json"
    },
    {
      "rfilename": "eval/2026-09-13/reference/tasks.bin"
    },
    {
      "rfilename": "eval/2026-09-13/reference/test.parquet"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-f32-run-manifest.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-f32-throughput-unvalidated.tsv"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-gpu01-parity.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-heldout-tokenizer-parity.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-middle-complete.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-native-cache-format-probe.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-oracle-few-vs-ref-f32kv.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-oracle-few-vs-ref-nofa.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-oracle-long-f32kv-boundary.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-quant-topology-2gpu-vs-4gpu.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-reference-f32-batching.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-reference-f32-short-comparison.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-reference-f32-short-nofa-comparison.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-shard-merge.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-shard-plan.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-tail-complete.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-topk-validation.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-vocab-parity.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek3.0-f32.gpu01-parity.tsv"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek3.0-f32.gpu0123-parity.tsv"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek3.0-f32.initial.tsv"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek3.0-f32.middle.tsv"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek3.0-f32.primary.tsv"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek3.0-f32.tail.tsv"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek3.0-f32.tsv"
    },
    {
      "rfilename": "eval/2026-09-13/results/diagnostics/deepseek3.0-f32.generation-probes.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/diagnostics/deepseek3.0-f32.metadata.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/diagnostics/deepseek3.0-f32.mini.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/diagnostics/deepseek3.0-f32.mini.native-specials.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/diagnostics/deepseek3.0-f32.ppl.log"
    },
    {
      "rfilename": "eval/2026-09-13/results/diagnostics/deepseek3.0-f32.ppl.validation.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/diagnostics/deepseek3.0-f32.request-plan.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/diagnostics/deepseek3.0-f32.serial-parity.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/diagnostics/deepseek3.0-f32.serial-parity.log"
    },
    {
      "rfilename": "eval/2026-09-13/results/diagnostics/deepseek3.0-f32.server0.log"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek-analysis-config.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek-analysis.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek-analysis.md"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek-f32-run-manifest.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek-f32-throughput-unvalidated.tsv"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek-gpu01-parity.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek-middle-complete.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek-shard-merge.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek-shard-plan.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek-tail-complete.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek3.0-f32.gpu01-parity.tsv"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek3.0-f32.initial.tsv"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek3.0-f32.middle.tsv"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek3.0-f32.primary.tsv"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek3.0-f32.tail.tsv"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek3.0-f32.tsv"
    },
    {
      "rfilename": "eval/2026-09-13/test_analyze_mmlu.py"
    },
    {
      "rfilename": "eval/2026-09-13/test_compare_ppl_caches.py"
    },
    {
      "rfilename": "runtime/LICENSE.DeepSeek"
    },
    {
      "rfilename": "runtime/LICENSE.llama.cpp"
    },
    {
      "rfilename": "runtime/README.md"
    },
    {
      "rfilename": "runtime/REPRODUCE.md"
    },
    {
      "rfilename": "runtime/build_helpers.sh"
    },
    {
      "rfilename": "runtime/capture.cpp"
    },
    {
      "rfilename": "runtime/deepseek-components.log"
    },
    {
      "rfilename": "runtime/deepseek-f32-throughput.log"
    },
    {
      "rfilename": "runtime/deepseek-gpu01-parity.json"
    },
    {
      "rfilename": "runtime/deepseek-heldout-tokenizer-parity.json"
    },
    {
      "rfilename": "runtime/deepseek-helper-build-check.log"
    },
    {
      "rfilename": "runtime/deepseek-native-vocab.bin"
    },
    {
      "rfilename": "runtime/deepseek-perplexity-sidecar.patch"
    },
    {
      "rfilename": "runtime/deepseek-preflight.log"
    },
    {
      "rfilename": "runtime/deepseek-quant-topology-2gpu-vs-4gpu.json"
    },
    {
      "rfilename": "runtime/deepseek-runtime-base-source.tar.gz"
    },
    {
      "rfilename": "runtime/deepseek-runtime.patch"
    },
    {
      "rfilename": "runtime/deepseek-shard-merge.json"
    },
    {
      "rfilename": "runtime/deepseek-topk-validation.json"
    },
    {
      "rfilename": "runtime/deepseek-vocab-parity.json"
    },
    {
      "rfilename": "runtime/deepseek3.0-f32.gpu0123-parity.tsv"
    },
    {
      "rfilename": "runtime/deepseek3.0-f32.tsv"
    },
    {
      "rfilename": "runtime/diagnostics/deepseek3.0-f32.generation-probes.json"
    },
    {
      "rfilename": "runtime/diagnostics/deepseek3.0-f32.metadata.json"
    },
    {
      "rfilename": "runtime/diagnostics/deepseek3.0-f32.mini.json"
    },
    {
      "rfilename": "runtime/diagnostics/deepseek3.0-f32.mini.native-specials.json"
    },
    {
      "rfilename": "runtime/diagnostics/deepseek3.0-f32.ppl.log"
    },
    {
      "rfilename": "runtime/diagnostics/deepseek3.0-f32.ppl.validation.json"
    },
    {
      "rfilename": "runtime/diagnostics/deepseek3.0-f32.request-plan.json"
    },
    {
      "rfilename": "runtime/diagnostics/deepseek3.0-f32.serial-parity.json"
    },
    {
      "rfilename": "runtime/diagnostics/deepseek3.0-f32.serial-parity.log"
    },
    {
      "rfilename": "runtime/diagnostics/deepseek3.0-f32.server0.log"
    },
    {
      "rfilename": "runtime/dump_vocab.cpp"
    },
    {
      "rfilename": "runtime/estimate_mmlu_time.py"
    },
    {
      "rfilename": "runtime/helper-build-validation.json"
    },
    {
      "rfilename": "runtime/mc-score-deepseek-f32.cpp"
    },
    {
      "rfilename": "runtime/nltk-resources.json"
    },
    {
      "rfilename": "runtime/nltk_data/tokenizers/punkt/README"
    },
    {
      "rfilename": "runtime/nltk_data/tokenizers/punkt/english.pickle"
    },
    {
      "rfilename": "runtime/nltk_data/tokenizers/punkt_tab/README"
    },
    {
      "rfilename": "runtime/nltk_data/tokenizers/punkt_tab/english/abbrev_types.txt"
    },
    {
      "rfilename": "runtime/nltk_data/tokenizers/punkt_tab/english/collocations.tab"
    },
    {
      "rfilename": "runtime/nltk_data/tokenizers/punkt_tab/english/ortho_context.tab"
    },
    {
      "rfilename": "runtime/nltk_data/tokenizers/punkt_tab/english/sent_starters.txt"
    },
    {
      "rfilename": "runtime/official-source/LICENSE"
    },
    {
      "rfilename": "runtime/official-source/config.json"
    },
    {
      "rfilename": "runtime/official-source/encoding/encoding.py"
    },
    {
      "rfilename": "runtime/official-source/inference/config.json"
    },
    {
      "rfilename": "runtime/official-source/inference/engram.py"
    },
    {
      "rfilename": "runtime/official-source/inference/image_processor.py"
    },
    {
      "rfilename": "runtime/official-source/inference/kernel.py"
    },
    {
      "rfilename": "runtime/official-source/inference/model.py"
    },
    {
      "rfilename": "runtime/official-source/inference/vision.py"
    },
    {
      "rfilename": "runtime/official-source/manifest.json"
    },
    {
      "rfilename": "runtime/official-source/tokenizer.json"
    },
    {
      "rfilename": "runtime/oracle/README.md"
    },
    {
      "rfilename": "runtime/oracle/cache_arithmetic.py"
    },
    {
      "rfilename": "runtime/oracle/compare_logits.py"
    },
    {
      "rfilename": "runtime/oracle/gguf_oracle.py"
    },
    {
      "rfilename": "runtime/oracle/results/deepseek-capture-few-ref-f32kv/logits.f32"
    },
    {
      "rfilename": "runtime/oracle/results/deepseek-capture-few-ref-f32kv/tokens.json"
    },
    {
      "rfilename": "runtime/oracle/results/deepseek-native-cache-format-probe.json"
    },
    {
      "rfilename": "runtime/oracle/results/deepseek-oracle-few-f32kv-cascade/logits.f32"
    },
    {
      "rfilename": "runtime/oracle/results/deepseek-oracle-few-f32kv-cascade/report.json"
    },
    {
      "rfilename": "runtime/oracle/results/deepseek-oracle-few-vs-ref-f32kv.json"
    },
    {
      "rfilename": "runtime/oracle/results/deepseek-oracle-long-f32kv-boundary.json"
    },
    {
      "rfilename": "runtime/oracle/results/deepseek-oracle-long-ref-f32kv-attention/report.json"
    },
    {
      "rfilename": "runtime/oracle/results/few-attention-f32-f16kv.json"
    },
    {
      "rfilename": "runtime/oracle/results/long-attention-f32-f16kv.json"
    },
    {
      "rfilename": "runtime/oracle/results/short-components-default.json"
    },
    {
      "rfilename": "runtime/oracle/results/short-components-f32.json"
    },
    {
      "rfilename": "runtime/perplexity.cpp"
    },
    {
      "rfilename": "runtime/portable-layout-validation.json"
    },
    {
      "rfilename": "runtime/prepare_replay_workspace.py"
    },
    {
      "rfilename": "runtime/provenance.json"
    },
    {
      "rfilename": "runtime/python-environment.json"
    },
    {
      "rfilename": "runtime/reference/deepseek-metadata.json"
    },
    {
      "rfilename": "runtime/reference/deepseek41.cpp"
    },
    {
      "rfilename": "runtime/reference/encoding-files.json"
    },
    {
      "rfilename": "runtime/reference/engram.py"
    },
    {
      "rfilename": "runtime/reference/model.py"
    },
    {
      "rfilename": "runtime/reference/subset.json"
    },
    {
      "rfilename": "runtime/reference/task-index.json"
    },
    {
      "rfilename": "runtime/reference/tasks.bin"
    },
    {
      "rfilename": "runtime/requirements-environment.txt"
    },
    {
      "rfilename": "runtime/requirements-minimal.txt"
    },
    {
      "rfilename": "runtime/rescore_diagnostics.py"
    },
    {
      "rfilename": "runtime/research/README.md"
    },
    {
      "rfilename": "runtime/research/deployment-crosscheck.md"
    },
    {
      "rfilename": "runtime/research/engine-implementations.md"
    },
    {
      "rfilename": "runtime/research/jigsaw_report.json"
    },
    {
      "rfilename": "runtime/research/jigsaw_runtime.json"
    },
    {
      "rfilename": "runtime/research/llama_conversion_pr.json"
    },
    {
      "rfilename": "runtime/research/miaai_runtime.json"
    },
    {
      "rfilename": "runtime/research/official-runtime.md"
    },
    {
      "rfilename": "runtime/research/question-state-isolation.md"
    },
    {
      "rfilename": "runtime/research/sglang_runtime_pr.json"
    },
    {
      "rfilename": "runtime/research/shi3z_runtime.json"
    },
    {
      "rfilename": "runtime/research/sources/DeepSeek_V41_Tech_Report.pdf"
    },
    {
      "rfilename": "runtime/research/sources/DeepSeek_V41_Tech_Report.txt"
    },
    {
      "rfilename": "runtime/research/sources/LICENSE"
    },
    {
      "rfilename": "runtime/research/sources/README.md"
    },
    {
      "rfilename": "runtime/research/sources/encoding/README.md"
    },
    {
      "rfilename": "runtime/research/sources/encoding_README.md"
    },
    {
      "rfilename": "runtime/research/sources/evaluation/README.md"
    },
    {
      "rfilename": "runtime/research/sources/evaluation_README.md"
    },
    {
      "rfilename": "runtime/research/sources/inference/README.md"
    },
    {
      "rfilename": "runtime/research/sources/inference_README.md"
    },
    {
      "rfilename": "runtime/research/sources/inference_config.json"
    },
    {
      "rfilename": "runtime/research/sources/inference_engram.py"
    },
    {
      "rfilename": "runtime/research/sources/inference_kernel.py"
    },
    {
      "rfilename": "runtime/research/sources/inference_model.py"
    },
    {
      "rfilename": "runtime/research/sources/manifest.json"
    },
    {
      "rfilename": "runtime/research/vcruz305_runtime.json"
    },
    {
      "rfilename": "runtime/run_diagnostics.executed.py"
    },
    {
      "rfilename": "runtime/run_diagnostics.py"
    },
    {
      "rfilename": "runtime/run_sharded_mmlu.py"
    },
    {
      "rfilename": "runtime/run_topology_gate.py"
    },
    {
      "rfilename": "runtime/test_reference_components.py"
    },
    {
      "rfilename": "runtime/validate_ppl.py"
    },
    {
      "rfilename": "runtime/validate_topk_masks.py"
    },
    {
      "rfilename": "runtime/verify_vocab.py"
    }
  ],
  "tags": [
    "gguf",
    "gsq",
    "rco",
    "quantization",
    "mixed-precision",
    "multimodal",
    "vision",
    "image-text-to-text",
    "arxiv:2604.18556",
    "arxiv:2605.00649",
    "base_model:deepseek-ai/DeepSeek-V4.1-Flash",
    "base_model:quantized:deepseek-ai/DeepSeek-V4.1-Flash",
    "doi:10.57967/hf/10402",
    "license:mit",
    "endpoints_compatible",
    "region:us",
    "conversational"
  ]
}
licencemit
receipt
Source
GGUF quantisations
Its words
mit
Read by
field:cardData.license
Said since
2026-09-29 09:50 UTC
Last answered
2026-10-04 18:16 UTC
Original
open at the source
What the source handed over
{
  "_id": "6aa69da6e29d415b0da7eee6",
  "author": "pfeifferj",
  "cardData": {
    "base_model": "deepseek-ai/DeepSeek-V4.1-Flash",
    "base_model_relation": "quantized",
    "library_name": "gguf",
    "license": "mit",
    "pipeline_tag": "image-text-to-text",
    "tags": [
      "gguf",
      "gsq",
      "rco",
      "quantization",
      "mixed-precision",
      "multimodal",
      "vision"
    ]
  },
  "createdAt": "2026-09-13T12:57:10.000Z",
  "downloads": 9685,
  "gated": false,
  "id": "pfeifferj/DeepSeek-V4.1-Flash-GSQ-RCO-GGUF",
  "lastModified": "2026-09-14T20:13:49.000Z",
  "library_name": "gguf",
  "likes": 12,
  "modelId": "pfeifferj/DeepSeek-V4.1-Flash-GSQ-RCO-GGUF",
  "pipeline_tag": "image-text-to-text",
  "private": false,
  "sha": "8270aceb26eb5590ea40f347256a1736999f551d",
  "siblings": [
    {
      "rfilename": ".gitattributes"
    },
    {
      "rfilename": "DeepSeek-V4.1-Flash-DSpark-draft.gguf"
    },
    {
      "rfilename": "DeepSeek-V4.1-Flash-GSQ-RCO-3.0bit.gguf"
    },
    {
      "rfilename": "DeepSeek-V4.1-Flash-mmproj-BF16.gguf"
    },
    {
      "rfilename": "LICENSE"
    },
    {
      "rfilename": "README.md"
    },
    {
      "rfilename": "SHA256SUMS"
    },
    {
      "rfilename": "eval/2026-09-13/PPL-COMPARISON.md"
    },
    {
      "rfilename": "eval/2026-09-13/README.md"
    },
    {
      "rfilename": "eval/2026-09-13/analyze_mmlu.py"
    },
    {
      "rfilename": "eval/2026-09-13/bench/NOTICE.md"
    },
    {
      "rfilename": "eval/2026-09-13/bench/evaluate_gguf.py"
    },
    {
      "rfilename": "eval/2026-09-13/bench/instruction_following_eval/LICENSE"
    },
    {
      "rfilename": "eval/2026-09-13/bench/instruction_following_eval/evaluation_lib.py"
    },
    {
      "rfilename": "eval/2026-09-13/bench/instruction_following_eval/instructions.py"
    },
    {
      "rfilename": "eval/2026-09-13/bench/instruction_following_eval/instructions_registry.py"
    },
    {
      "rfilename": "eval/2026-09-13/bench/instruction_following_eval/instructions_util.py"
    },
    {
      "rfilename": "eval/2026-09-13/bench/mini_bench.py"
    },
    {
      "rfilename": "eval/2026-09-13/bench/request-plan.json"
    },
    {
      "rfilename": "eval/2026-09-13/bench/run_diagnostics.py"
    },
    {
      "rfilename": "eval/2026-09-13/bench/run_ppl.py"
    },
    {
      "rfilename": "eval/2026-09-13/bench/summarize_diagnostics.py"
    },
    {
      "rfilename": "eval/2026-09-13/compare_ppl_caches.py"
    },
    {
      "rfilename": "eval/2026-09-13/data/calibration-manifest.json"
    },
    {
      "rfilename": "eval/2026-09-13/data/download-manifest.json"
    },
    {
      "rfilename": "eval/2026-09-13/data/gsm8k.json"
    },
    {
      "rfilename": "eval/2026-09-13/data/heldout.txt"
    },
    {
      "rfilename": "eval/2026-09-13/data/ifeval.jsonl"
    },
    {
      "rfilename": "eval/2026-09-13/diagnostics/README.md"
    },
    {
      "rfilename": "eval/2026-09-13/diagnostics/check_scorer.py"
    },
    {
      "rfilename": "eval/2026-09-13/diagnostics/mini_bench_glm_adapter.py"
    },
    {
      "rfilename": "eval/2026-09-13/diagnostics/run_diagnostics.py"
    },
    {
      "rfilename": "eval/2026-09-13/diagnostics/summarize_diagnostics.py"
    },
    {
      "rfilename": "eval/2026-09-13/provenance/deepseek-mmlu-runtime.json"
    },
    {
      "rfilename": "eval/2026-09-13/reference/build_tasks.py"
    },
    {
      "rfilename": "eval/2026-09-13/reference/mc-score.cpp"
    },
    {
      "rfilename": "eval/2026-09-13/reference/questions.json"
    },
    {
      "rfilename": "eval/2026-09-13/reference/subset.json"
    },
    {
      "rfilename": "eval/2026-09-13/reference/task-index.json"
    },
    {
      "rfilename": "eval/2026-09-13/reference/tasks.bin"
    },
    {
      "rfilename": "eval/2026-09-13/reference/test.parquet"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-f32-run-manifest.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-f32-throughput-unvalidated.tsv"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-gpu01-parity.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-heldout-tokenizer-parity.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-middle-complete.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-native-cache-format-probe.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-oracle-few-vs-ref-f32kv.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-oracle-few-vs-ref-nofa.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-oracle-long-f32kv-boundary.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-quant-topology-2gpu-vs-4gpu.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-reference-f32-batching.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-reference-f32-short-comparison.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-reference-f32-short-nofa-comparison.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-shard-merge.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-shard-plan.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-tail-complete.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-topk-validation.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-vocab-parity.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek3.0-f32.gpu01-parity.tsv"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek3.0-f32.gpu0123-parity.tsv"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek3.0-f32.initial.tsv"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek3.0-f32.middle.tsv"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek3.0-f32.primary.tsv"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek3.0-f32.tail.tsv"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek3.0-f32.tsv"
    },
    {
      "rfilename": "eval/2026-09-13/results/diagnostics/deepseek3.0-f32.generation-probes.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/diagnostics/deepseek3.0-f32.metadata.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/diagnostics/deepseek3.0-f32.mini.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/diagnostics/deepseek3.0-f32.mini.native-specials.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/diagnostics/deepseek3.0-f32.ppl.log"
    },
    {
      "rfilename": "eval/2026-09-13/results/diagnostics/deepseek3.0-f32.ppl.validation.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/diagnostics/deepseek3.0-f32.request-plan.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/diagnostics/deepseek3.0-f32.serial-parity.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/diagnostics/deepseek3.0-f32.serial-parity.log"
    },
    {
      "rfilename": "eval/2026-09-13/results/diagnostics/deepseek3.0-f32.server0.log"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek-analysis-config.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek-analysis.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek-analysis.md"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek-f32-run-manifest.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek-f32-throughput-unvalidated.tsv"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek-gpu01-parity.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek-middle-complete.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek-shard-merge.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek-shard-plan.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek-tail-complete.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek3.0-f32.gpu01-parity.tsv"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek3.0-f32.initial.tsv"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek3.0-f32.middle.tsv"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek3.0-f32.primary.tsv"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek3.0-f32.tail.tsv"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek3.0-f32.tsv"
    },
    {
      "rfilename": "eval/2026-09-13/test_analyze_mmlu.py"
    },
    {
      "rfilename": "eval/2026-09-13/test_compare_ppl_caches.py"
    },
    {
      "rfilename": "runtime/LICENSE.DeepSeek"
    },
    {
      "rfilename": "runtime/LICENSE.llama.cpp"
    },
    {
      "rfilename": "runtime/README.md"
    },
    {
      "rfilename": "runtime/REPRODUCE.md"
    },
    {
      "rfilename": "runtime/build_helpers.sh"
    },
    {
      "rfilename": "runtime/capture.cpp"
    },
    {
      "rfilename": "runtime/deepseek-components.log"
    },
    {
      "rfilename": "runtime/deepseek-f32-throughput.log"
    },
    {
      "rfilename": "runtime/deepseek-gpu01-parity.json"
    },
    {
      "rfilename": "runtime/deepseek-heldout-tokenizer-parity.json"
    },
    {
      "rfilename": "runtime/deepseek-helper-build-check.log"
    },
    {
      "rfilename": "runtime/deepseek-native-vocab.bin"
    },
    {
      "rfilename": "runtime/deepseek-perplexity-sidecar.patch"
    },
    {
      "rfilename": "runtime/deepseek-preflight.log"
    },
    {
      "rfilename": "runtime/deepseek-quant-topology-2gpu-vs-4gpu.json"
    },
    {
      "rfilename": "runtime/deepseek-runtime-base-source.tar.gz"
    },
    {
      "rfilename": "runtime/deepseek-runtime.patch"
    },
    {
      "rfilename": "runtime/deepseek-shard-merge.json"
    },
    {
      "rfilename": "runtime/deepseek-topk-validation.json"
    },
    {
      "rfilename": "runtime/deepseek-vocab-parity.json"
    },
    {
      "rfilename": "runtime/deepseek3.0-f32.gpu0123-parity.tsv"
    },
    {
      "rfilename": "runtime/deepseek3.0-f32.tsv"
    },
    {
      "rfilename": "runtime/diagnostics/deepseek3.0-f32.generation-probes.json"
    },
    {
      "rfilename": "runtime/diagnostics/deepseek3.0-f32.metadata.json"
    },
    {
      "rfilename": "runtime/diagnostics/deepseek3.0-f32.mini.json"
    },
    {
      "rfilename": "runtime/diagnostics/deepseek3.0-f32.mini.native-specials.json"
    },
    {
      "rfilename": "runtime/diagnostics/deepseek3.0-f32.ppl.log"
    },
    {
      "rfilename": "runtime/diagnostics/deepseek3.0-f32.ppl.validation.json"
    },
    {
      "rfilename": "runtime/diagnostics/deepseek3.0-f32.request-plan.json"
    },
    {
      "rfilename": "runtime/diagnostics/deepseek3.0-f32.serial-parity.json"
    },
    {
      "rfilename": "runtime/diagnostics/deepseek3.0-f32.serial-parity.log"
    },
    {
      "rfilename": "runtime/diagnostics/deepseek3.0-f32.server0.log"
    },
    {
      "rfilename": "runtime/dump_vocab.cpp"
    },
    {
      "rfilename": "runtime/estimate_mmlu_time.py"
    },
    {
      "rfilename": "runtime/helper-build-validation.json"
    },
    {
      "rfilename": "runtime/mc-score-deepseek-f32.cpp"
    },
    {
      "rfilename": "runtime/nltk-resources.json"
    },
    {
      "rfilename": "runtime/nltk_data/tokenizers/punkt/README"
    },
    {
      "rfilename": "runtime/nltk_data/tokenizers/punkt/english.pickle"
    },
    {
      "rfilename": "runtime/nltk_data/tokenizers/punkt_tab/README"
    },
    {
      "rfilename": "runtime/nltk_data/tokenizers/punkt_tab/english/abbrev_types.txt"
    },
    {
      "rfilename": "runtime/nltk_data/tokenizers/punkt_tab/english/collocations.tab"
    },
    {
      "rfilename": "runtime/nltk_data/tokenizers/punkt_tab/english/ortho_context.tab"
    },
    {
      "rfilename": "runtime/nltk_data/tokenizers/punkt_tab/english/sent_starters.txt"
    },
    {
      "rfilename": "runtime/official-source/LICENSE"
    },
    {
      "rfilename": "runtime/official-source/config.json"
    },
    {
      "rfilename": "runtime/official-source/encoding/encoding.py"
    },
    {
      "rfilename": "runtime/official-source/inference/config.json"
    },
    {
      "rfilename": "runtime/official-source/inference/engram.py"
    },
    {
      "rfilename": "runtime/official-source/inference/image_processor.py"
    },
    {
      "rfilename": "runtime/official-source/inference/kernel.py"
    },
    {
      "rfilename": "runtime/official-source/inference/model.py"
    },
    {
      "rfilename": "runtime/official-source/inference/vision.py"
    },
    {
      "rfilename": "runtime/official-source/manifest.json"
    },
    {
      "rfilename": "runtime/official-source/tokenizer.json"
    },
    {
      "rfilename": "runtime/oracle/README.md"
    },
    {
      "rfilename": "runtime/oracle/cache_arithmetic.py"
    },
    {
      "rfilename": "runtime/oracle/compare_logits.py"
    },
    {
      "rfilename": "runtime/oracle/gguf_oracle.py"
    },
    {
      "rfilename": "runtime/oracle/results/deepseek-capture-few-ref-f32kv/logits.f32"
    },
    {
      "rfilename": "runtime/oracle/results/deepseek-capture-few-ref-f32kv/tokens.json"
    },
    {
      "rfilename": "runtime/oracle/results/deepseek-native-cache-format-probe.json"
    },
    {
      "rfilename": "runtime/oracle/results/deepseek-oracle-few-f32kv-cascade/logits.f32"
    },
    {
      "rfilename": "runtime/oracle/results/deepseek-oracle-few-f32kv-cascade/report.json"
    },
    {
      "rfilename": "runtime/oracle/results/deepseek-oracle-few-vs-ref-f32kv.json"
    },
    {
      "rfilename": "runtime/oracle/results/deepseek-oracle-long-f32kv-boundary.json"
    },
    {
      "rfilename": "runtime/oracle/results/deepseek-oracle-long-ref-f32kv-attention/report.json"
    },
    {
      "rfilename": "runtime/oracle/results/few-attention-f32-f16kv.json"
    },
    {
      "rfilename": "runtime/oracle/results/long-attention-f32-f16kv.json"
    },
    {
      "rfilename": "runtime/oracle/results/short-components-default.json"
    },
    {
      "rfilename": "runtime/oracle/results/short-components-f32.json"
    },
    {
      "rfilename": "runtime/perplexity.cpp"
    },
    {
      "rfilename": "runtime/portable-layout-validation.json"
    },
    {
      "rfilename": "runtime/prepare_replay_workspace.py"
    },
    {
      "rfilename": "runtime/provenance.json"
    },
    {
      "rfilename": "runtime/python-environment.json"
    },
    {
      "rfilename": "runtime/reference/deepseek-metadata.json"
    },
    {
      "rfilename": "runtime/reference/deepseek41.cpp"
    },
    {
      "rfilename": "runtime/reference/encoding-files.json"
    },
    {
      "rfilename": "runtime/reference/engram.py"
    },
    {
      "rfilename": "runtime/reference/model.py"
    },
    {
      "rfilename": "runtime/reference/subset.json"
    },
    {
      "rfilename": "runtime/reference/task-index.json"
    },
    {
      "rfilename": "runtime/reference/tasks.bin"
    },
    {
      "rfilename": "runtime/requirements-environment.txt"
    },
    {
      "rfilename": "runtime/requirements-minimal.txt"
    },
    {
      "rfilename": "runtime/rescore_diagnostics.py"
    },
    {
      "rfilename": "runtime/research/README.md"
    },
    {
      "rfilename": "runtime/research/deployment-crosscheck.md"
    },
    {
      "rfilename": "runtime/research/engine-implementations.md"
    },
    {
      "rfilename": "runtime/research/jigsaw_report.json"
    },
    {
      "rfilename": "runtime/research/jigsaw_runtime.json"
    },
    {
      "rfilename": "runtime/research/llama_conversion_pr.json"
    },
    {
      "rfilename": "runtime/research/miaai_runtime.json"
    },
    {
      "rfilename": "runtime/research/official-runtime.md"
    },
    {
      "rfilename": "runtime/research/question-state-isolation.md"
    },
    {
      "rfilename": "runtime/research/sglang_runtime_pr.json"
    },
    {
      "rfilename": "runtime/research/shi3z_runtime.json"
    },
    {
      "rfilename": "runtime/research/sources/DeepSeek_V41_Tech_Report.pdf"
    },
    {
      "rfilename": "runtime/research/sources/DeepSeek_V41_Tech_Report.txt"
    },
    {
      "rfilename": "runtime/research/sources/LICENSE"
    },
    {
      "rfilename": "runtime/research/sources/README.md"
    },
    {
      "rfilename": "runtime/research/sources/encoding/README.md"
    },
    {
      "rfilename": "runtime/research/sources/encoding_README.md"
    },
    {
      "rfilename": "runtime/research/sources/evaluation/README.md"
    },
    {
      "rfilename": "runtime/research/sources/evaluation_README.md"
    },
    {
      "rfilename": "runtime/research/sources/inference/README.md"
    },
    {
      "rfilename": "runtime/research/sources/inference_README.md"
    },
    {
      "rfilename": "runtime/research/sources/inference_config.json"
    },
    {
      "rfilename": "runtime/research/sources/inference_engram.py"
    },
    {
      "rfilename": "runtime/research/sources/inference_kernel.py"
    },
    {
      "rfilename": "runtime/research/sources/inference_model.py"
    },
    {
      "rfilename": "runtime/research/sources/manifest.json"
    },
    {
      "rfilename": "runtime/research/vcruz305_runtime.json"
    },
    {
      "rfilename": "runtime/run_diagnostics.executed.py"
    },
    {
      "rfilename": "runtime/run_diagnostics.py"
    },
    {
      "rfilename": "runtime/run_sharded_mmlu.py"
    },
    {
      "rfilename": "runtime/run_topology_gate.py"
    },
    {
      "rfilename": "runtime/test_reference_components.py"
    },
    {
      "rfilename": "runtime/validate_ppl.py"
    },
    {
      "rfilename": "runtime/validate_topk_masks.py"
    },
    {
      "rfilename": "runtime/verify_vocab.py"
    }
  ],
  "tags": [
    "gguf",
    "gsq",
    "rco",
    "quantization",
    "mixed-precision",
    "multimodal",
    "vision",
    "image-text-to-text",
    "arxiv:2604.18556",
    "arxiv:2605.00649",
    "base_model:deepseek-ai/DeepSeek-V4.1-Flash",
    "base_model:quantized:deepseek-ai/DeepSeek-V4.1-Flash",
    "doi:10.57967/hf/10402",
    "license:mit",
    "endpoints_compatible",
    "region:us",
    "conversational"
  ]
}
likes12
receipt
Source
GGUF quantisations
Its words
12
Read by
field:likes
Said since
2026-09-29 09:50 UTC
Last answered
2026-10-04 18:16 UTC
Original
open at the source
What the source handed over
{
  "_id": "6aa69da6e29d415b0da7eee6",
  "author": "pfeifferj",
  "cardData": {
    "base_model": "deepseek-ai/DeepSeek-V4.1-Flash",
    "base_model_relation": "quantized",
    "library_name": "gguf",
    "license": "mit",
    "pipeline_tag": "image-text-to-text",
    "tags": [
      "gguf",
      "gsq",
      "rco",
      "quantization",
      "mixed-precision",
      "multimodal",
      "vision"
    ]
  },
  "createdAt": "2026-09-13T12:57:10.000Z",
  "downloads": 9685,
  "gated": false,
  "id": "pfeifferj/DeepSeek-V4.1-Flash-GSQ-RCO-GGUF",
  "lastModified": "2026-09-14T20:13:49.000Z",
  "library_name": "gguf",
  "likes": 12,
  "modelId": "pfeifferj/DeepSeek-V4.1-Flash-GSQ-RCO-GGUF",
  "pipeline_tag": "image-text-to-text",
  "private": false,
  "sha": "8270aceb26eb5590ea40f347256a1736999f551d",
  "siblings": [
    {
      "rfilename": ".gitattributes"
    },
    {
      "rfilename": "DeepSeek-V4.1-Flash-DSpark-draft.gguf"
    },
    {
      "rfilename": "DeepSeek-V4.1-Flash-GSQ-RCO-3.0bit.gguf"
    },
    {
      "rfilename": "DeepSeek-V4.1-Flash-mmproj-BF16.gguf"
    },
    {
      "rfilename": "LICENSE"
    },
    {
      "rfilename": "README.md"
    },
    {
      "rfilename": "SHA256SUMS"
    },
    {
      "rfilename": "eval/2026-09-13/PPL-COMPARISON.md"
    },
    {
      "rfilename": "eval/2026-09-13/README.md"
    },
    {
      "rfilename": "eval/2026-09-13/analyze_mmlu.py"
    },
    {
      "rfilename": "eval/2026-09-13/bench/NOTICE.md"
    },
    {
      "rfilename": "eval/2026-09-13/bench/evaluate_gguf.py"
    },
    {
      "rfilename": "eval/2026-09-13/bench/instruction_following_eval/LICENSE"
    },
    {
      "rfilename": "eval/2026-09-13/bench/instruction_following_eval/evaluation_lib.py"
    },
    {
      "rfilename": "eval/2026-09-13/bench/instruction_following_eval/instructions.py"
    },
    {
      "rfilename": "eval/2026-09-13/bench/instruction_following_eval/instructions_registry.py"
    },
    {
      "rfilename": "eval/2026-09-13/bench/instruction_following_eval/instructions_util.py"
    },
    {
      "rfilename": "eval/2026-09-13/bench/mini_bench.py"
    },
    {
      "rfilename": "eval/2026-09-13/bench/request-plan.json"
    },
    {
      "rfilename": "eval/2026-09-13/bench/run_diagnostics.py"
    },
    {
      "rfilename": "eval/2026-09-13/bench/run_ppl.py"
    },
    {
      "rfilename": "eval/2026-09-13/bench/summarize_diagnostics.py"
    },
    {
      "rfilename": "eval/2026-09-13/compare_ppl_caches.py"
    },
    {
      "rfilename": "eval/2026-09-13/data/calibration-manifest.json"
    },
    {
      "rfilename": "eval/2026-09-13/data/download-manifest.json"
    },
    {
      "rfilename": "eval/2026-09-13/data/gsm8k.json"
    },
    {
      "rfilename": "eval/2026-09-13/data/heldout.txt"
    },
    {
      "rfilename": "eval/2026-09-13/data/ifeval.jsonl"
    },
    {
      "rfilename": "eval/2026-09-13/diagnostics/README.md"
    },
    {
      "rfilename": "eval/2026-09-13/diagnostics/check_scorer.py"
    },
    {
      "rfilename": "eval/2026-09-13/diagnostics/mini_bench_glm_adapter.py"
    },
    {
      "rfilename": "eval/2026-09-13/diagnostics/run_diagnostics.py"
    },
    {
      "rfilename": "eval/2026-09-13/diagnostics/summarize_diagnostics.py"
    },
    {
      "rfilename": "eval/2026-09-13/provenance/deepseek-mmlu-runtime.json"
    },
    {
      "rfilename": "eval/2026-09-13/reference/build_tasks.py"
    },
    {
      "rfilename": "eval/2026-09-13/reference/mc-score.cpp"
    },
    {
      "rfilename": "eval/2026-09-13/reference/questions.json"
    },
    {
      "rfilename": "eval/2026-09-13/reference/subset.json"
    },
    {
      "rfilename": "eval/2026-09-13/reference/task-index.json"
    },
    {
      "rfilename": "eval/2026-09-13/reference/tasks.bin"
    },
    {
      "rfilename": "eval/2026-09-13/reference/test.parquet"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-f32-run-manifest.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-f32-throughput-unvalidated.tsv"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-gpu01-parity.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-heldout-tokenizer-parity.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-middle-complete.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-native-cache-format-probe.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-oracle-few-vs-ref-f32kv.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-oracle-few-vs-ref-nofa.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-oracle-long-f32kv-boundary.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-quant-topology-2gpu-vs-4gpu.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-reference-f32-batching.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-reference-f32-short-comparison.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-reference-f32-short-nofa-comparison.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-shard-merge.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-shard-plan.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-tail-complete.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-topk-validation.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek-vocab-parity.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek3.0-f32.gpu01-parity.tsv"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek3.0-f32.gpu0123-parity.tsv"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek3.0-f32.initial.tsv"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek3.0-f32.middle.tsv"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek3.0-f32.primary.tsv"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek3.0-f32.tail.tsv"
    },
    {
      "rfilename": "eval/2026-09-13/results/deepseek3.0-f32.tsv"
    },
    {
      "rfilename": "eval/2026-09-13/results/diagnostics/deepseek3.0-f32.generation-probes.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/diagnostics/deepseek3.0-f32.metadata.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/diagnostics/deepseek3.0-f32.mini.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/diagnostics/deepseek3.0-f32.mini.native-specials.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/diagnostics/deepseek3.0-f32.ppl.log"
    },
    {
      "rfilename": "eval/2026-09-13/results/diagnostics/deepseek3.0-f32.ppl.validation.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/diagnostics/deepseek3.0-f32.request-plan.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/diagnostics/deepseek3.0-f32.serial-parity.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/diagnostics/deepseek3.0-f32.serial-parity.log"
    },
    {
      "rfilename": "eval/2026-09-13/results/diagnostics/deepseek3.0-f32.server0.log"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek-analysis-config.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek-analysis.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek-analysis.md"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek-f32-run-manifest.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek-f32-throughput-unvalidated.tsv"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek-gpu01-parity.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek-middle-complete.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek-shard-merge.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek-shard-plan.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek-tail-complete.json"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek3.0-f32.gpu01-parity.tsv"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek3.0-f32.initial.tsv"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek3.0-f32.middle.tsv"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek3.0-f32.primary.tsv"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek3.0-f32.tail.tsv"
    },
    {
      "rfilename": "eval/2026-09-13/results/mmlu/deepseek3.0-f32.tsv"
    },
    {
      "rfilename": "eval/2026-09-13/test_analyze_mmlu.py"
    },
    {
      "rfilename": "eval/2026-09-13/test_compare_ppl_caches.py"
    },
    {
      "rfilename": "runtime/LICENSE.DeepSeek"
    },
    {
      "rfilename": "runtime/LICENSE.llama.cpp"
    },
    {
      "rfilename": "runtime/README.md"
    },
    {
      "rfilename": "runtime/REPRODUCE.md"
    },
    {
      "rfilename": "runtime/build_helpers.sh"
    },
    {
      "rfilename": "runtime/capture.cpp"
    },
    {
      "rfilename": "runtime/deepseek-components.log"
    },
    {
      "rfilename": "runtime/deepseek-f32-throughput.log"
    },
    {
      "rfilename": "runtime/deepseek-gpu01-parity.json"
    },
    {
      "rfilename": "runtime/deepseek-heldout-tokenizer-parity.json"
    },
    {
      "rfilename": "runtime/deepseek-helper-build-check.log"
    },
    {
      "rfilename": "runtime/deepseek-native-vocab.bin"
    },
    {
      "rfilename": "runtime/deepseek-perplexity-sidecar.patch"
    },
    {
      "rfilename": "runtime/deepseek-preflight.log"
    },
    {
      "rfilename": "runtime/deepseek-quant-topology-2gpu-vs-4gpu.json"
    },
    {
      "rfilename": "runtime/deepseek-runtime-base-source.tar.gz"
    },
    {
      "rfilename": "runtime/deepseek-runtime.patch"
    },
    {
      "rfilename": "runtime/deepseek-shard-merge.json"
    },
    {
      "rfilename": "runtime/deepseek-topk-validation.json"
    },
    {
      "rfilename": "runtime/deepseek-vocab-parity.json"
    },
    {
      "rfilename": "runtime/deepseek3.0-f32.gpu0123-parity.tsv"
    },
    {
      "rfilename": "runtime/deepseek3.0-f32.tsv"
    },
    {
      "rfilename": "runtime/diagnostics/deepseek3.0-f32.generation-probes.json"
    },
    {
      "rfilename": "runtime/diagnostics/deepseek3.0-f32.metadata.json"
    },
    {
      "rfilename": "runtime/diagnostics/deepseek3.0-f32.mini.json"
    },
    {
      "rfilename": "runtime/diagnostics/deepseek3.0-f32.mini.native-specials.json"
    },
    {
      "rfilename": "runtime/diagnostics/deepseek3.0-f32.ppl.log"
    },
    {
      "rfilename": "runtime/diagnostics/deepseek3.0-f32.ppl.validation.json"
    },
    {
      "rfilename": "runtime/diagnostics/deepseek3.0-f32.request-plan.json"
    },
    {
      "rfilename": "runtime/diagnostics/deepseek3.0-f32.serial-parity.json"
    },
    {
      "rfilename": "runtime/diagnostics/deepseek3.0-f32.serial-parity.log"
    },
    {
      "rfilename": "runtime/diagnostics/deepseek3.0-f32.server0.log"
    },
    {
      "rfilename": "runtime/dump_vocab.cpp"
    },
    {
      "rfilename": "runtime/estimate_mmlu_time.py"
    },
    {
      "rfilename": "runtime/helper-build-validation.json"
    },
    {
      "rfilename": "runtime/mc-score-deepseek-f32.cpp"
    },
    {
      "rfilename": "runtime/nltk-resources.json"
    },
    {
      "rfilename": "runtime/nltk_data/tokenizers/punkt/README"
    },
    {
      "rfilename": "runtime/nltk_data/tokenizers/punkt/english.pickle"
    },
    {
      "rfilename": "runtime/nltk_data/tokenizers/punkt_tab/README"
    },
    {
      "rfilename": "runtime/nltk_data/tokenizers/punkt_tab/english/abbrev_types.txt"
    },
    {
      "rfilename": "runtime/nltk_data/tokenizers/punkt_tab/english/collocations.tab"
    },
    {
      "rfilename": "runtime/nltk_data/tokenizers/punkt_tab/english/ortho_context.tab"
    },
    {
      "rfilename": "runtime/nltk_data/tokenizers/punkt_tab/english/sent_starters.txt"
    },
    {
      "rfilename": "runtime/official-source/LICENSE"
    },
    {
      "rfilename": "runtime/official-source/config.json"
    },
    {
      "rfilename": "runtime/official-source/encoding/encoding.py"
    },
    {
      "rfilename": "runtime/official-source/inference/config.json"
    },
    {
      "rfilename": "runtime/official-source/inference/engram.py"
    },
    {
      "rfilename": "runtime/official-source/inference/image_processor.py"
    },
    {
      "rfilename": "runtime/official-source/inference/kernel.py"
    },
    {
      "rfilename": "runtime/official-source/inference/model.py"
    },
    {
      "rfilename": "runtime/official-source/inference/vision.py"
    },
    {
      "rfilename": "runtime/official-source/manifest.json"
    },
    {
      "rfilename": "runtime/official-source/tokenizer.json"
    },
    {
      "rfilename": "runtime/oracle/README.md"
    },
    {
      "rfilename": "runtime/oracle/cache_arithmetic.py"
    },
    {
      "rfilename": "runtime/oracle/compare_logits.py"
    },
    {
      "rfilename": "runtime/oracle/gguf_oracle.py"
    },
    {
      "rfilename": "runtime/oracle/results/deepseek-capture-few-ref-f32kv/logits.f32"
    },
    {
      "rfilename": "runtime/oracle/results/deepseek-capture-few-ref-f32kv/tokens.json"
    },
    {
      "rfilename": "runtime/oracle/results/deepseek-native-cache-format-probe.json"
    },
    {
      "rfilename": "runtime/oracle/results/deepseek-oracle-few-f32kv-cascade/logits.f32"
    },
    {
      "rfilename": "runtime/oracle/results/deepseek-oracle-few-f32kv-cascade/report.json"
    },
    {
      "rfilename": "runtime/oracle/results/deepseek-oracle-few-vs-ref-f32kv.json"
    },
    {
      "rfilename": "runtime/oracle/results/deepseek-oracle-long-f32kv-boundary.json"
    },
    {
      "rfilename": "runtime/oracle/results/deepseek-oracle-long-ref-f32kv-attention/report.json"
    },
    {
      "rfilename": "runtime/oracle/results/few-attention-f32-f16kv.json"
    },
    {
      "rfilename": "runtime/oracle/results/long-attention-f32-f16kv.json"
    },
    {
      "rfilename": "runtime/oracle/results/short-components-default.json"
    },
    {
      "rfilename": "runtime/oracle/results/short-components-f32.json"
    },
    {
      "rfilename": "runtime/perplexity.cpp"
    },
    {
      "rfilename": "runtime/portable-layout-validation.json"
    },
    {
      "rfilename": "runtime/prepare_replay_workspace.py"
    },
    {
      "rfilename": "runtime/provenance.json"
    },
    {
      "rfilename": "runtime/python-environment.json"
    },
    {
      "rfilename": "runtime/reference/deepseek-metadata.json"
    },
    {
      "rfilename": "runtime/reference/deepseek41.cpp"
    },
    {
      "rfilename": "runtime/reference/encoding-files.json"
    },
    {
      "rfilename": "runtime/reference/engram.py"
    },
    {
      "rfilename": "runtime/reference/model.py"
    },
    {
      "rfilename": "runtime/reference/subset.json"
    },
    {
      "rfilename": "runtime/reference/task-index.json"
    },
    {
      "rfilename": "runtime/reference/tasks.bin"
    },
    {
      "rfilename": "runtime/requirements-environment.txt"
    },
    {
      "rfilename": "runtime/requirements-minimal.txt"
    },
    {
      "rfilename": "runtime/rescore_diagnostics.py"
    },
    {
      "rfilename": "runtime/research/README.md"
    },
    {
      "rfilename": "runtime/research/deployment-crosscheck.md"
    },
    {
      "rfilename": "runtime/research/engine-implementations.md"
    },
    {
      "rfilename": "runtime/research/jigsaw_report.json"
    },
    {
      "rfilename": "runtime/research/jigsaw_runtime.json"
    },
    {
      "rfilename": "runtime/research/llama_conversion_pr.json"
    },
    {
      "rfilename": "runtime/research/miaai_runtime.json"
    },
    {
      "rfilename": "runtime/research/official-runtime.md"
    },
    {
      "rfilename": "runtime/research/question-state-isolation.md"
    },
    {
      "rfilename": "runtime/research/sglang_runtime_pr.json"
    },
    {
      "rfilename": "runtime/research/shi3z_runtime.json"
    },
    {
      "rfilename": "runtime/research/sources/DeepSeek_V41_Tech_Report.pdf"
    },
    {
      "rfilename": "runtime/research/sources/DeepSeek_V41_Tech_Report.txt"
    },
    {
      "rfilename": "runtime/research/sources/LICENSE"
    },
    {
      "rfilename": "runtime/research/sources/README.md"
    },
    {
      "rfilename": "runtime/research/sources/encoding/README.md"
    },
    {
      "rfilename": "runtime/research/sources/encoding_README.md"
    },
    {
      "rfilename": "runtime/research/sources/evaluation/README.md"
    },
    {
      "rfilename": "runtime/research/sources/evaluation_README.md"
    },
    {
      "rfilename": "runtime/research/sources/inference/README.md"
    },
    {
      "rfilename": "runtime/research/sources/inference_README.md"
    },
    {
      "rfilename": "runtime/research/sources/inference_config.json"
    },
    {
      "rfilename": "runtime/research/sources/inference_engram.py"
    },
    {
      "rfilename": "runtime/research/sources/inference_kernel.py"
    },
    {
      "rfilename": "runtime/research/sources/inference_model.py"
    },
    {
      "rfilename": "runtime/research/sources/manifest.json"
    },
    {
      "rfilename": "runtime/research/vcruz305_runtime.json"
    },
    {
      "rfilename": "runtime/run_diagnostics.executed.py"
    },
    {
      "rfilename": "runtime/run_diagnostics.py"
    },
    {
      "rfilename": "runtime/run_sharded_mmlu.py"
    },
    {
      "rfilename": "runtime/run_topology_gate.py"
    },
    {
      "rfilename": "runtime/test_reference_components.py"
    },
    {
      "rfilename": "runtime/validate_ppl.py"
    },
    {
      "rfilename": "runtime/validate_topk_masks.py"
    },
    {
      "rfilename": "runtime/verify_vocab.py"
    }
  ],
  "tags": [
    "gguf",
    "gsq",
    "rco",
    "quantization",
    "mixed-precision",
    "multimodal",
    "vision",
    "image-text-to-text",
    "arxiv:2604.18556",
    "arxiv:2605.00649",
    "base_model:deepseek-ai/DeepSeek-V4.1-Flash",
    "base_model:quantized:deepseek-ai/DeepSeek-V4.1-Flash",
    "doi:10.57967/hf/10402",
    "license:mit",
    "endpoints_compatible",
    "region:us",
    "conversational"
  ]
}

Text

This source's terms allow its title, its values and a link here, not its text. It is at https://huggingface.co/pfeifferj/DeepSeek-V4.1-Flash-GSQ-RCO-GGUF.