游戏/图形引擎 · C

antirez/ds4

DeepSeek 4 Flash and PRO local inference engine for Metal, CUDA and ROCm。

增长榜 #32 已读 GitHub / README
增长排名 #32 Fast Growth Top 100
本期热度 Stars 39 OSSInsight 页面展示
Forks 4 榜单记录
Fork / Star 10.3% 社区复用强度

项目解读

DeepSeek 4 Flash and PRO local inference engine for Metal, CUDA and ROCm。 README 重点章节包括:DwarfStar、Motivations、Acknowledgements to llama.cpp and GGML、Status、More Documentation。

README / GitHub 亮点

  • GitHub 描述:DeepSeek 4 Flash and PRO local inference engine for Metal, CUDA and ROCm。
  • DwarfStar is a small native inference engine optimized first for。
  • DeepSeek V4 Flash, with support for DeepSeek V4 PRO on very high-memory。
  • intentionally narrow: not a generic GGUF runner, not a wrapper around another。

适用场景

适合评估 AI 应用、智能体工作流、模型工具链、RAG/提示词工程或 AI 辅助开发场景。

采用前核查

采用前仍需核查许可证、维护节奏、issue 质量、release 记录和生产适配成本。

README 摘要

DwarfStar is a small native inference engine optimized first for DeepSeek V4 Flash, with support for DeepSeek V4 PRO on very high-memory intentionally narrow: not a generic GGUF runner, not a wrapper around another runtime: it is completely self-contained. Other than running the model in a correct and fast way, the project goal is to provide DeepSeek specific loading, prompt rendering, tool calling, KV state handling (RAM and on-disk), server