Skip to content

@LLM360 LLM360

All

    Repositories list

    56 repositories

    • miles

      Public
      Miles is an enterprise-facing reinforcement learning framework for LLM and VLM post-training, forked from and co-evolving with slime.
      Python
      •
      Apache License 2.0
      •564•0•3•18•Updated Oct 16, 2026Oct 16, 2026

    • mini_task_runner

      Public
      Shell
      •0•0•0•0•Updated Oct 15, 2026Oct 15, 2026

    • harbor

      Public
      Harbor is a framework for running agent evaluations and creating and using RL environments.
      Python
      •
      Apache License 2.0
      •2k•0•4•60•Updated Oct 15, 2026Oct 15, 2026

    • sglang

      Public
      SGLang is a high-performance serving framework for large language models and multimodal models.
      Python
      •
      Apache License 2.0
      •9.6k•0•4•20•Updated Oct 9, 2026Oct 9, 2026

    • lemonade

      Public
      Lemonade helps users discover and run local AI apps by serving optimized LLMs right from their own GPUs and NPUs. Join our discord: https://discord.gg/5xXzkMu8Z…
      C++
      •
      Apache License 2.0
      •529•0•2•0•Updated Oct 8, 2026Oct 8, 2026

    • foreman-xllm-public-fork

      Public
      Light-weight infrastructure for express LLM pre-training
      Python
      •
      Apache License 2.0
      •12•0•0•0•Updated Oct 5, 2026Oct 5, 2026

    • Megatron-LM

      Public
      Ongoing research training transformer models at scale
      Python
      •
      Other
      •4.6k•0•0•3•Updated Sep 18, 2026Sep 18, 2026

    • smg

      Public
      Engine-agnostic LLM gateway in Rust. Full OpenAI & Anthropic API compatibility across SGLang, vLLM, TRT-LLM, OpenAI, Gemini & more. Industry-first gRPC pipeline…
      Rust
      •
      Apache License 2.0
      •188•1•20•8•Updated Sep 10, 2026Sep 10, 2026

    • scientific-agent-skills

      Public
      Turn any AI agent into an AI Scientist. The #1 Agent Skills library for science, used by 160,000+ scientists worldwide. 140 ready-to-use skills plus 100+ scient…
      Python
      •
      MIT License
      •4.5k•0•0•0•Updated Sep 2, 2026Sep 2, 2026

    • swe-atlas-qna

      Public
      Python
      •0•0•0•0•Updated Sep 1, 2026Sep 1, 2026

    • swe-bench-pro

      Public
      Shell
      •0•0•0•0•Updated Sep 1, 2026Sep 1, 2026

    • pier

      Public
      Pier is a Harbor fork built for DeepSWE, with stronger support for CLI agents in air-gapped (no-internet) tasks and more faithful, consistent agent trajectories
      Python
      •
      Apache License 2.0
      •58•0•0•1•Updated Sep 1, 2026Sep 1, 2026

    • terminal-bench-2

      Public
      Shell
      •
      Apache License 2.0
      •117•0•0•0•Updated Jul 31, 2026Jul 31, 2026

    • ProgramBench

      Public
      Can Language Models Rebuild Programs From Scratch?
      Python
      •
      MIT License
      •67•0•0•0•Updated Jul 29, 2026Jul 29, 2026

    • vllm

      Public
      A high-throughput and memory-efficient inference and serving engine for LLMs
      Python
      •
      Apache License 2.0
      •23k•0•0•0•Updated Jul 23, 2026Jul 23, 2026

    • archipelago

      Public
      Harness for running and evaluating AI agents against RL environments
      Python
      •
      Apache License 2.0
      •69•0•0•0•Updated Jul 22, 2026Jul 22, 2026

    • gorilla

      Public
      Gorilla: Training and Evaluating LLMs for Function Calls (Tool Calls)
      Python
      •
      Apache License 2.0
      •1.4k•0•0•0•Updated Jul 22, 2026Jul 22, 2026

    • mcpmark

      Public
      MCPMark is a comprehensive, stress-testing MCP benchmark designed to evaluate model and agent capabilities in real-world MCP use.
      Python
      •
      Apache License 2.0
      •50•0•0•0•Updated Jul 18, 2026Jul 18, 2026

    • EdgeBench

      Public
      EdgeBench: Unveiling scaling laws of learning from real-world environments
      Python
      •
      Apache License 2.0
      •20•0•0•0•Updated Jul 17, 2026Jul 17, 2026

    • terminal-bench-2-1

      Public
      Terminal-Bench 2.1
      Shell
      •
      Apache License 2.0
      •66•0•0•0•Updated Jul 14, 2026Jul 14, 2026

    • WildClawBench

      Public
      An in-the-wild benchmark for AI agents in the OpenClaw Environment.
      Python
      •
      MIT License
      •65•0•0•0•Updated Jul 13, 2026Jul 13, 2026

    • rBridge

      Public
      TeX
      •1•0•0•0•Updated Jul 8, 2026Jul 8, 2026

    • Eval360

      Public
      Shell
      •
      Apache License 2.0
      •3•11•0•0•Updated Jul 7, 2026Jul 7, 2026

    • Toolathlon

      Public
      [ICLR 2026] The Tool Decathlon: Benchmarking Language Agents for Diverse, Realistic, and Long-Horizon Task Execution
      Python
      •70•0•0•0•Updated Jul 6, 2026Jul 6, 2026

    • Reasoning360

      Public
      A repo for open research on building large reasoning models
      rlreasoningr1
      rlreasoningr1llmqwen
      Python
      •
      Apache License 2.0
      •19•154•1•13•Updated Jul 3, 2026Jul 3, 2026

    • BrowseComp-Plus

      Public
      BrowseComp-Plus: A More Fair and Transparent Evaluation Benchmark of Deep-Research Agent (ACL 2026 Main)
      Python
      •
      MIT License
      •68•0•0•0•Updated Jul 2, 2026Jul 2, 2026

    • simple-evals-browsecomp

      Public
      Python
      •
      MIT License
      •515•0•0•0•Updated Jun 24, 2026Jun 24, 2026

    • router

      Public
      A high-performance and light-weight router for vLLM large scale deployment
      Rust
      •
      Apache License 2.0
      •219•0•0•1•Updated Jun 22, 2026Jun 22, 2026

    • eval360-snapshots

      Public
      Dashboard for tracking LLM evaluation results across checkpoints and datasets
      HTML
      •0•0•0•0•Updated Jun 19, 2026Jun 19, 2026

    • PostTrainBench

      Public
      Measuring how well CLI agents like Claude Code or Codex CLI can post-train base LLMs on a single H100 GPU in 10 hours
      Python
      •
      MIT License
      •68•0•0•0•Updated Jun 4, 2026Jun 4, 2026
    ProTip! When viewing an organization's repositories, you can use the props. filter to filter by custom property.