open-compass/opencompass

opencompass

✅ Recently maintained

OpenCompass is an LLM evaluation platform, supporting a wide range of models from OpenAI, Anthropic, Gemini, Qwen, GLM, DeepSeek, etc, across 100+ datasets covering knowledge, reasoning, coding, science, language, long-context, and safety.

7,561Stars
900Forks
+127-day stars

Python · Apache-2.0 · Last push 2026-11-02

Official repository & installation ↗
evaluationbenchmarklarge-language-modelllmllm-eval

Star history

2025-09-292026-11-02
View historical data
DateStars
2026-10-047,478
2026-10-057,479
2026-10-067,481
2026-10-077,483
2026-10-087,488
2026-10-097,493
2026-10-107,496
2026-10-117,496
2026-10-127,498
2026-10-137,503
2026-10-147,509
2026-10-157,511
2026-10-167,512
2026-10-177,515
2026-10-187,519
2026-10-197,521
2026-10-207,523
2026-10-217,527
2026-10-227,530
2026-10-237,534
2026-10-247,539
2026-10-257,544
2026-10-267,549
2026-10-277,552
2026-10-287,556
2026-10-297,556
2026-10-307,559
2026-10-317,559
2026-11-017,560
2026-11-027,561

Get the project

git clone https://github.com/open-compass/opencompass.git

Related projects & alternative candidates

Related domains do not imply feature parity. Review official documentation and licenses.

promptfoo

Test your prompts, agents, and RAGs. Red teaming/pentesting/vulnerability scanning for AI. Compare performance of GPT, Claude, Gemini, DeepSeek, and more. Simpl

★ 26,723hermes-agent

Open-source AI agent from Nous Research that works with Claude, GPT and other models and grows with use.

★ 259,826firecrawl

Crawls and scrapes web pages into clean Markdown for AI agents and LLM pipelines.

★ 197,975