Repository Index

A high-throughput and memory-efficient inference and serving engine for LLMs

• gpt
• llm
• pytorch
• model-serving

Local UI to run and train LLMs and diffusion models. Supports GGUF, MLX, Qwen3.8, DeepSeek-V4.1, MiniMax-H3, Gemma 4, FLUX and more.

• fine-tuning
• llama
• llms
• gemma

Qwen3 is the large language model series developed by Qwen team, Alibaba Cloud.

Translated🔥 MaxKB is an open-source platform for building enterprise-grade agents. A powerful and easy-to-use open-source enterprise-grade agent platform.

• llm
• ollama
• maxkb
• knowledgebase

Qwen3-Coder is the code version of Qwen3, the large language model series developed by Qwen team.

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).

• llm
• lora
• sft
• multimodal

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!

• llms
• lora
• reinforcement-learning
• agent

Jev-like family of decision models built on top of Qwen3.5/3.8 you can train and run on your own

• decision-model
• jev
• qwen3

~95% on SimpleQA (e.g. Qwen3.6-27B on a 3090). Supports all local and cloud LLMs (llama.cpp, Ollama, Google, ...). 10+ search engines - arXiv, PubMed, your private documents. Everything Local & Encrypted.

• academia
• arxiv
• brave
• deep-research

Open Source Deep Research Alternative to Reason and Search on Private Data. Written in Python.

• agent
• llm
• rag
• vector-database

A Next-Generation Training Engine Built for Ultra-Large MoE Models

• llm
• agent
• deepseek-v3
• gpt-oss

Semantic search over videos using Gemini Embedding 2 or Qwen3-VL.

• chromadb
• dashcam
• gemini
• gemini-embedding-2

Steering-first neuron explorer for Qwen3.5-4B: push every MLP neuron, keep every rollout, browse the results

• interpretability
• mechanistic-interpretability
• qwen
• steering

GVS5H: Five Qwen3.8-27B Models Match Claude Fable 5 on LiveCodeBench Hard | Fable 5 Level Coding for a Fifth the Price - or on a Single GPU

Run 284B DeepSeek-V4-Flash (<59GiB RAM: 41.1 cold pp, 9.1 N=1 tg, up to 25.3 spec tok/s) & 30B MoEs (22.4 N=1 / 111.6 spec tok/s) on 64GB CPU (0 GPUs).

• avx512
• cpu-inference
• hpc
• llama-cpp

WeVisDoc is an end-to-end document parser for page images. Fine-tuned from Qwen3-VL-2B-Instruct and Qwen3-VL-4B-Instruct, it turns a page into structured Markdown, with LaTeX formulas and HTML tables.

JevStyle: local decision models for AI agents. Classify, route and score with a probability per option. systemone-compatible local server, MCP tools, Claude Code guard. Open weights, Apache-2.0.

• agent-skills
• calibration
• claude-code
• decision-model