Awesome Llm Apps
Automation & Agents100+ AI Agents, Agent Skills and RAG Apps - Free and Open Source.
Free
ClawBench: review, pricing & alternatives
ClawBench is an open-source benchmark that evaluates browser AI agents on everyday online tasks. It provides a live leaderboard ranking agents across 130 real-world tasks using a two-stage scoring system (HTTP-request interception and LLM judge).
What is ClawBench?
ClawBench is an open-source benchmark that evaluates browser AI agents on everyday online tasks. It provides a live leaderboard ranking agents across 130 real-world tasks using a two-stage scoring system (HTTP-request interception and LLM judge).
Best for
- Benchmarking browser-based AI agents
- Comparing model performance on real-world online tasks
- Researching agent capabilities across diverse task sets
- Evaluating new AI agents before deployment
ClawBench features
- Open-source benchmark for browser AI agents
- Live leaderboard ranking on 130 real-world tasks
- Two-stage scoring: HTTP-request interception + LLM judge
- claw-eval: 300 human-verified tasks across 9 categories (Pass³ scoring)
- InternLM/WildClawBench: 60 in-the-wild tasks in a live OpenClaw environment
- ClawMark: 100 multi-day coworker-agent tasks with deterministic Python checkers
- Free to use as listed on the official site
ClawBench pricing
The official site lists the product as "Free"; no paid tiers are mentioned.
Price lines as shown on the official site
- 6% $4.4425 58 / 130 2 gpt-5
- 5% $0.3325 46 / 130 3 glm-5
- 7% $0.1935 45 / 130 4 deepseek-v4-pro hermes 33
- 3% $0.0721 44 / 130 5 deepseek-v4-flash:free hermes 2
- 0% $0.0000 3 / 130 6 z-ai/glm-4
- 8% $0.0000 3 / 130 7 minimax-m2
Sources: official homepage · last checked 2026-10-27. Prices can change; confirm on the official site.
How to use ClawBench
- Visit the ClawBench website and open the dashboard.
- Select the AI agent or model you want to evaluate.
- Run the benchmark; the system executes the chosen tasks and records HTTP requests.
- View the two-stage scores and the agent's ranking on the live leaderboard.
- Analyze the results to compare performance across models or task categories.
ClawBench pros and cons
Pros (from the official site)
- Open-source and freely available
- Comprehensive task sets (human-verified, in-the-wild, multi-day)
- Live leaderboard for transparent ranking
- Two-stage scoring provides detailed performance insight
Limits stated on the site
None stated on the official site.
ClawBench alternatives
Browser Use
Automation & AgentsAgents that use the browser.
Free
Ragflow
Automation & AgentsRAGFlow is a leading open-source Retrieval-Augmented Generation (RAG) engine that fuses cutting-edge RAG with Agent capabilities to create a superior context layer for LLMs
Free
OpenHands
Automation & Agents🙌 OpenHands: AI-Driven Development
Free
Lobehub
Automation & Agents🤯 LobeHub is your Chief Agent Operator, organizing your agents into 7×24 operations by hiring, scheduling, and reporting on your entire AI team.
Free
Hello Agents
Automation & Agents📚 《从零开始构建智能体》——从零开始的智能体原理与实践教程
Free
ClawBench FAQ
What is ClawBench?
ClawBench is an open-source benchmark that evaluates browser AI agents on everyday online tasks. It provides a live leaderboard ranking agents across 130 real-world tasks using a two-stage scoring system (HTTP-request interception and LLM judge).
Is ClawBench free?
Yes. ClawBench's official website mentions a free plan, free tier or trial (checked 2026-10-27).
How much does ClawBench cost?
The official site lists the product as "Free"; no paid tiers are mentioned.
How do I use ClawBench?
1. Visit the ClawBench website and open the dashboard. 2. Select the AI agent or model you want to evaluate. 3. Run the benchmark; the system executes the chosen tasks and records HTTP requests.
What are the best alternatives to ClawBench?
Alternatives in Automation & Agents include Awesome Llm Apps, Browser Use, Ragflow, OpenHands, Lobehub.
Written from ClawBench's official website (checked 2026-10-27); nothing here is sponsored. Want to master tools like this? See the 100+ AI Tools Course.