Pinned Loading

1. VLMSAM VLMSAM Public

   Qwen-SAM is a reasoning-based segmentation model that integrates Qwen 2.5 VL 7B with the Segment Anything Model (SAM), enabling fine-grained visual segmentation from complex text prompts using LoRA…

   Python 20 1

2. RLCHF RLCHF Public

   RLCHF provides tools to generate diverse persona-based preference data and train/evaluate alignment models using RLHF/DPO workflows. It includes persona generation scripts, example persona datasets…

   Jupyter Notebook

3. LLada-Reasoning LLada-Reasoning Public

   LLaDA-Reasoning: A long-context reasoning and diffusion-style model (8K context, dual KV cache) built with an end-to-end pipeline that includes pretraining, supervised fine-tuning (SFT), DPO + GRPO…

   Python

4. CadQuery-Code-Generator CadQuery-Code-Generator Public

   This project creates a CadQuery code generator model by fine-tuning vision-language models (Internvl 3).

   Python

5. adrsimon/gomagnon adrsimon/gomagnon Public

   A Go multi-agent ecosystem of prehistoric humans trying to evolve as a society.

   Go 1