Free
Free
The benchmark site, leaderboard, and dataset access are public with no paid subscription listed.
- Access
- The benchmark site, leaderboard, and dataset access are public with no paid subscription listed.
ClawBench
● Identity checkedResearch · Vendor site 2026-10-04
ClawBench is an open benchmark and public leaderboard for AI browser agents on live websites. It scores models on 130 real-world tasks across dozens of platforms using HTTP interception plus an LLM judge, with datasets and open submissions via pull request.
- Category
- Research
- Free access
- The benchmark site, leaderboard, and dataset access are public with no paid subscription listed.
- API access
- Not confirmed as a commercial API; the project publishes tasks, traces, and open submission instructions on the site.
Is ClawBench right for you?
A good fit for
Researchers and agent developers comparing browser automation models on realistic checkout, booking, and form-filling tasks.
Before you choose
Reported per-task costs on the leaderboard reflect model inference spend during evaluation, not a ClawBench subscription fee.
Live browser-agent benchmark
ClawBench ranks AI agents on everyday online tasks on real sites, publishing rewards, interception rates, and pass counts for public comparison.
What it can do
Features & capabilities
Unknown is different from unavailable. Each fact carries its own evidence.
| Capability | Value | Evidence | Checked |
|---|---|---|---|
| Task corpus | Homepage lists 130 tasks across 63 live platforms such as booking flights and ordering food. | Facts sourced | 2026-10-04 |
| Scoring method | Benchmark uses two-stage scoring combining HTTP-request interception and an LLM judge. | Facts sourced | 2026-10-04 |
| Open submissions | Site invites open model submissions via pull request alongside arXiv and Hugging Face dataset links. | Facts sourced | 2026-10-04 |
Understand the total cost
ClawBench pricing & plans
Alternatives to ClawBench
View all ↗
AllyhubAllyhub provides AI agents that collect, monitor, and act on web data from social platforms, marketplaces, and the open web. Users describe what to track, and agents gather fresh data, watch for meaningful changes, and deliver scheduled insights.
Explore tool
OrimonOrimon.ai is a no-code platform for building generative AI chatbots that automate support, qualify leads, and answer internal questions across websites, WhatsApp, Instagram, and other channels. It advertises multilingual support, analytics, live-agent handoff, and integrations such as HubSpot.
Explore tool
ChatBotKitChatBotKit is an AI agent platform for support, sales, and operations teams to launch widgets and messaging bots trained on company data. It supports Slack, Discord, WhatsApp, Telegram, enterprise search, and API-first deployments with observability and security features.
Explore toolThe practical questions
Frequently asked questions
What is ClawBench used for?
ClawBench is an open benchmark and public leaderboard for AI browser agents on live websites. It scores models on 130 real-world tasks across dozens of platforms using HTTP interception plus an LLM judge, with datasets and open submissions via pull request.
Does ClawBench have a free plan?
The benchmark site, leaderboard, and dataset access are public with no paid subscription listed.. This record lists ongoing free access; check the plan limits before starting.
How much does ClawBench cost?
No paid monthly price is listed; this record treats the product as free to start. See the plan cards for entitlements, billing commitments and seat minimums.
Can I use ClawBench through an API?
Not confirmed as a commercial API; the project publishes tasks, traces, and open submission instructions on the site.. API access and subscription access may have different terms; consult the linked sources.
What should I check before choosing it?
Reported per-task costs on the leaderboard reflect model inference spend during evaluation, not a ClawBench subscription fee.
Price history
No retained pricing changes yet. A current price alone does not establish a historical trend.
How this profile is supported
Identity and profile
Reviewed vendor source
Read original source ↗Facts apply to the named version and check date. Send a sourced correction if something changed.