samuel-vitorino/lm.rs

• highreadme#1

  Reposition the README's introductory text to emphasize functional capability

  原因：

  当前

  Inspired by Karpathy's llama2.c and llm.c I decided to create the most minimal code (not so minimal atm) that can perform full inference on Language Models on the CPU without ML libraries. Previously only Google's Gemma 2 models were supported, but I decided to add support for the new Llama 3.2 models, and more recently the option to use images with PHI-3.5.

  复制粘贴的修复

  复制

  Inspired by Karpathy's `llama2.c` and `llm.c`, `lm.rs` provides a minimal, self-contained Rust implementation for full LLM inference on the CPU, without external ML libraries. It supports models like Gemma 2, Llama 3.2, and PHI-3.5 (including multimodal vision). While designed for clarity and direct CPU execution, it's actively being optimized for performance, with recent updates boosting batch processing for image encoding.

  待办完成忽略
• hightopics#2

  Add relevant topics to the repository

  原因：

  复制粘贴的修复

  复制

  llm, inference, rust, cpu, multimodal, language-models, machine-learning, edge-ai, phi-3-5, llama-3

  待办完成忽略

本次扫描解析到的品类 GEO 通道：google/gemini-2.5-flash, deepseek/deepseek-v4-flash

• 品类问题

  Looking for a lightweight Rust library to perform local LLM inference without GPU.

  你：未被推荐
  AI 推荐顺序：

  1. llm crate (huggingface/llm)
  2. candle (huggingface/candle)
  3. rust-bert (huggingface/rust-bert)
  4. tract (sonos/tract)
  5. ort (pykeio/ort)

  AI 推荐了 5 个替代方案，却始终没点名 samuel-vitorino/lm.rs。这就是要补上的差距。

  查看 AI 完整回答
• 品类问题

  Seeking a Rust solution for multimodal LLM inference on edge devices.

  你：未被推荐
  AI 推荐顺序：

  1. ONNX Runtime
  2. OpenVINO
  3. Apache TVM
  4. candle (huggingface/candle)
  5. tract (sonos/tract)

  AI 推荐了 5 个替代方案，却始终没点名 samuel-vitorino/lm.rs。这就是要补上的差距。

  查看 AI 完整回答

• Metadata completeness

  warn

  建议：

• README presence

  pass

• Compared to common alternatives in this category, what is the core differentiator of samuel-vitorino/lm.rs?

  pass
  ⚠AI 未点名 samuel-vitorino/lm.rs —— 很可能在说另一个项目

  AI 的回答可能信誓旦旦却是错的。请按事实核对：技术栈、目标人群、差异化点是不是和你实际的对得上？

• If a team adopts samuel-vitorino/lm.rs in production, what risks or prerequisites should they evaluate first?

  pass
  ✓AI 明确点名了 samuel-vitorino/lm.rs

  AI 的回答可能信誓旦旦却是错的。请按事实核对：技术栈、目标人群、差异化点是不是和你实际的对得上？

• In one sentence, what problem does the repo samuel-vitorino/lm.rs solve, and who is the primary audience?

  pass
  ✓AI 明确点名了 samuel-vitorino/lm.rs

  AI 的回答可能信誓旦旦却是错的。请按事实核对：技术栈、目标人群、差异化点是不是和你实际的对得上？