jaredpalmer/kev

• Diamond III

Non-autoregressive System 1 decision engine. Typed choice, score and yes/no decisions over any text in a single forward pass, in 100+ languages, with a router that picks the right checkpoint per request.

• 2,014−12 Elo, change today
• 35.5k stars
• 3.1k forks
• 123 watchers
• 5h ago

git clone https://github.com/NandhaKishorM/laya && cd laya && docker compose up -d

• calibration
• classification
• decision-model
• huggingface

• Platinum II

Use PEFT or Full-parameter to CPT/SFT/DPO/GRPO 600+ LLMs (Qwen3.8, DeepSeek-V4, GLM-5.1, InternLM3, Llama4, ...) and 300+ MLLMs (Qwen3-VL, Qwen3-Omni, InternVL3.5, Ovis2.5, GLM5.3, Gemma4, Llava, Phi4, ...) (AAAI 2025).

• 1,708−11 Elo, change today
• 16.1k stars
• 1.8k forks
• 55 watchers
• 11h ago

git clone https://github.com/modelscope/ms-swift.git

• deepseek
• embedding
• glm
• grpo

• Silver IV

Native MLX runtime for Laya typed decision models — 7–14 ms short decisions on M3 Max. No text generation, PyTorch, or cloud API.

• 1,457−4 Elo, change today
• 7.1k stars
• 576 forks
• 17d ago

git clone https://github.com/mizorewww/laya-mlx.git

• apple-silicon
• decision-model
• inference
• laya

• Silver II

Agent Reinforcement Trainer: train multi-step agents for real-world tasks using GRPO. Give your agents on-the-job training. Reinforcement learning for Qwen3.6, GPT-OSS, Llama, and more!

• 1,493−6 Elo, change today
• 10.8k stars
• 1k forks
• 20h ago

git clone https://github.com/OpenPipe/ART.git

• agent
• agentic-ai
• grpo
• llms