hiyouga/LlamaFactory
Unified Efficient Fine-Tuning of 100+ LLMs & VLMs (ACL 2024)
This period is measured from GitHub’s rebuilt star history — one point per week — because our own daily snapshots do not reach this far back yet. Period boundaries can be off by up to half a week; the repository page prints the real start date for each period. So far the history has been fetched for 12,569 of 63,708 tracked repositories, and only those can be ranked in this period.
Star counts measured between 09:43 and 00:55 (BKK).
Unified Efficient Fine-Tuning of 100+ LLMs & VLMs (ACL 2024)
Faster Whisper transcription with CTranslate2
Serve large Qwen models fast on the GPUs you actually own. Qwen3.8-27B on a single 24 GB card with vLLM: 127 tok/s single-user (381 when the answer quotes the prompt), ~1,035 tok/s at 64 concurrent, 150k-262k context. vLLM patches, requant pipeline, benchmarks.
Janas-LLM: large mixture-of-experts language models on ordinary computers, in C. Experts streamed from NVMe, with or without a GPU.
A vector index built on TurboQuant, written in Rust with Python bindings
RDNA-native LLM inference engine in Rust.
State-of-the-art LLM compression, built for production inference with vLLM
Halo is an open-source framework built by White Circle for training large language and multimodal models
llama.cpp fork for significantly improved performance on Ampere (especially RTX 3090 / 3090 Ti): example: 95+ tok/s over a 100K-token generation at temperature 1 for qwen3.8, supporting full 262K ctx. Supports low memory (12gb) configs for 200k+ ctx
A unified inference runtime for VLA models.
Build, train, optimize, and run computer vision models locally, from raw images to live inference. Open source, optimized for Intel® XPU (CUDA also supported).
🚀 Accelerate inference and training of 🤗 Transformers, Diffusers, TIMM and Sentence Transformers with easy to use hardware optimization tools
Pure-Rust, CPU-only OCR engine for Baidu Unlimited-OCR (a DeepSeek-OCR-derived 3B MoE VLM). Five-model zoo, custom int8 kernels, no ML framework, no Python, no GPU.
Dataflow compiler for QNN inference on FPGAs
That's all 14 repositories in this period.