Stars
Official Luth-2 repository for French SFT, RL, and evaluation.
Luth is a state-of-the-art series of fine-tuned LLMs for French
Academic Research Skills for Claude Code: research → write → review → revise → finalize
Reinforcement Learning via Self-Distillation (SDPO)
NextPlaid, ColGREP: Multi-vector search, from database to coding agents.
Open-source evaluation toolkit of large multi-modality models (LMMs), support 220+ LMMs, 80+ benchmarks
A free modern, minimalist portfolio template built with Astro. Features a collection of blog posts, projects, and professional experiences.
Provider-agnostic, open-source evaluation infrastructure for language models
Tools for merging pretrained large language models.
Machine Learning Engineering Open Book
A high-throughput and memory-efficient inference and serving engine for LLMs
docTR (Document Text Recognition) - a seamless, high-performing & accessible library for OCR-related tasks powered by Deep Learning. Ongoing development and maintenance by t2k.
The scripts used to run the evaluation pipeline to generate fr-gouv-coordination-ia/llm_leaderboard_fr
A framework for few-shot evaluation of language models.
Hypernetworks that adapt LLMs for specific benchmark tasks using only textual task description as the input
An open-source tool for LLM prompt optimization.
Prompts for our Grok chat assistant and the `@grok` bot on X.
[ICLR 2026] Tina: Tiny Reasoning Models via LoRA
Fully open reproduction of DeepSeek-R1
250+ Fine-tuning & RL Notebooks for text, vision, audio, embedding, TTS models.
High-Performance Engine for Multi-Vector Search
Parallel Scaling Law for Language Model — Beyond Parameter and Inference Time Scaling