bestofgithub.dev

← Back to the directory

HKUDS/DeepCode

View on GitHub

"DeepCode: Open Agentic Coding (Agent Harness & Loop Engineering & Multi-Agent Orchestration)"

HKUDS/DeepCode is #2,985 on the bestofgithub Elo ladder, in Platinum V, with a rating of 1,638 (−47 this week). It has 16,845 stars, 21 gained in the last week.

Platinum V

1,638 Elo

  • #2,985 on the ladder
  • −18 today · −47 this week
  • Peak 1,690
9 Elo to Platinum IV

Stars

16,845

  • +0 today
  • +21 this week
  • – this month

Repository

Forks
2.1k
Language
Python
Last push

Achievements

bestofgithub badgeMaintainer? Show your league in your README

Card: your league's emblem, from Copper to Legendary

[![bestofgithub](https://bestofgithub.dev/badge/HKUDS/DeepCode.svg?style=card)](https://bestofgithub.dev/repo/HKUDS/DeepCode)

Flat: fits beside your other badges

[![bestofgithub](https://bestofgithub.dev/badge/HKUDS/DeepCode.svg)](https://bestofgithub.dev/repo/HKUDS/DeepCode)

Install

git
git clone https://github.com/HKUDS/DeepCode.git

Similar repositories

letta-ai/letta

Open on GitHub

Platform for stateful agents: AI with advanced memory that can learn and self-improve over time.

  • 1,800+5 Elo, change today
  • 25.4k stars
  • 2.6k forks
  • 142 watchers
git
git clone https://github.com/letta-ai/letta.git
0.16.86mo ago
  • ai
  • ai-agents
  • llm
  • llm-agent

revfactory/harness

Open on GitHub

A meta-skill that designs domain-specific agent teams, defines specialized agents, and generates the skills they use.

  • 1,591+4 Elo, change today
  • 9.3k stars
  • 1.3k forks
git
git clone https://github.com/revfactory/harness.git
  • claude-code
  • claude-code-plugin
  • harness
  • harness-engineering

truefoundry/trueforge

Open on GitHubTypeScript

The open-source agent harness - the runtime layer that turns an LLM into a working agent.

  • 1,503+5 Elo, change today
  • 6.4k stars
  • 561 forks
git
git clone https://github.com/truefoundry/trueforge.git
  • agent
  • agentic-ai
  • agents
  • harness

jeinlee1991/chinese-llm-benchmark

Open on GitHub

非线智能 NoneLinear - ReLE评测：中文AI大模型能力评测（持续更新）：目前已囊括374个大模型，覆盖chatgpt、gpt-5.4、谷歌gemini-3.1-pro、Claude-4.6、文心ERNIE-X1.1、ERNIE-5.0、qwen3.6-max、qwen3.6-plus、百川、讯飞星火、商汤senseChat等商用模型， 以及step3.5-flash、kimi-k2.6、ernie4.5、MiniMax-M2.7、deepseek-v4、Qwen3.6、llama4、智谱GLM-5.1、MiMo-V2、LongCat、gemma4、mistral等开源大模型。不仅提供排行榜，也提供规模超200万的大模型缺陷库！方便广大社区研究分析、改进大模型。

  • 1,331+5 Elo, change today
  • 6.5k stars
  • 261 forks
git
git clone https://github.com/jeinlee1991/chinese-llm-benchmark.git
  • agentic-ai
  • artificial-intelligence
  • llm-agent
  • llm-evaluation

pydantic/pydantic-ai

Open on GitHubPython

How Python does AI. Agents, realtime voice, image generation, embeddings. Every model, every interface, typed end to end.

  • 1,904+4 Elo, change today
  • 21.2k stars
  • 3.1k forks
  • 117 watchers
git
git clone https://github.com/pydantic/pydantic-ai.git
v2.68.03d ago
  • agent-framework
  • genai
  • harness
  • harness-engineering

README

Loading README…