📦 转存说明

• 来源仓库：unsloth/DeepSeek-V4-Flash-0731-GGUF
• 本仓库用途：将上述源仓库中的部分文件转存至此，目录结构与源仓库保持一致。
• 转存文件：
  • UD-Q8_K_XL/ — DeepSeek-V4-Flash-0731 的 UD-Q8_K_XL 量化模型（5 个 gguf 分片）
    • DeepSeek-V4-Flash-0731-UD-Q8_K_XL-00001-of-00005.gguf ~ -00005-of-00005.gguf
  • dspark/ — DSpark 相关文件
    • dspark-DeepSeek-V4-Flash-0731-BF16.gguf
    • README.md
  • dspark-DeepSeek-V4-Flash-0731-Q8_0.gguf（根目录的 Q8_0 drafter）

以下为源仓库的原始 README 说明：

Read our How to Run DeepSeek-V4-0731 Guide!

• To run DeepSeek-V4-Flash-0731 in full precision lossless, run Q8 (UD-Q8_K_XL), which is 162GB and only 7GB bigger than Q4 (UD-Q4_K_XL).
• See our DeepSeek-V4 guide for quantization analysis and instructions.
• You can now run DeepSeek-V4-Flash-0731 in Unsloth Studio with toggles for High and Max thinking.

DeepSeek-V4-Flash-0731

Introduction

1. For the Code Agent tasks among the public benchmarks above, DeepSeek-V4-Flash-0731 is evaluated with the minimal mode of DeepSeek Harness (to be released) as the agent framework, using the max reasoning effort level with temperature = 1.0, top_p = 0.95.
2. † DSBench-FullStack is an internal full-stack development test set; DSBench-Hard is an internal test set of difficult coding-agent problems.

License

Citation

@misc{deepseekai2026deepseekv4,
      title={DeepSeek-V4: Towards Highly Efficient Million-Token Context Intelligence},
      author={DeepSeek-AI},
      year={2026},
}