Unified Efficient Fine-Tuning of 100+ LLMs & VLMs (ACL 2024)

1 min voice data can also be used to train a good TTS model! (few shot voice cloning)

gpt-oss-120b and gpt-oss-20b are two open-weight language models by OpenAI

A high-performance, zero-overhead, extensible Python compiler with built-in NumPy support

A computer algebra system written in pure Python

Scalene: a high-performance, high-precision CPU, GPU, and memory profiler for Python with AI-powered optimization proposals

Use PEFT or Full-parameter to CPT/SFT/DPO/GRPO 500+ LLMs (Qwen3, Qwen3-MoE, Llama4, GLM4.5, InternLM3, DeepSeek-R1, ...) and 200+ MLLMs (Qwen3-VL, Qwen3-Omni, InternVL3.5, Ovis2.5, Llava, GLM4v, Ph…

LMDeploy is a toolkit for compressing, deploying, and serving LLMs.

Lark is a parsing toolkit for Python, built with a focus on ergonomics, performance and modularity.

Official Implementation of EAGLE-1 (ICML'24), EAGLE-2 (EMNLP'24), and EAGLE-3 (NeurIPS'25).

Command line utility for forced alignment using Kaldi

OpenAI-Compatible RESTful APIs for Amazon Bedrock