Artificial Analysis
Models
Third-party benchmark charts for every model Artificial Analysis tracks, with the ones rtxclaw has enabled marked on each chart.
—Models tracked
—rtxclaw enabled
—Index version
—API tier
—Requests left
—Source
Artificial Analysis API key required
- Create a free key on the Artificial Analysis login page (Free tier, internal use with attribution).
-
Append it to the operator env file, then reload this tab:
echo 'ARTIFICIAL_ANALYSIS_API_KEY=your-key-here' >> ~/.rtxclaw/.env
The web server resolves the key from its own environment,
~/.rtxclaw/.env, or /etc/environment. The key is
never sent to the browser and never written by this page.
rtxclaw enabled
enabled via an approximate match (≈)
not enabled in rtxclaw
running locally (rtxclaw serves it here)
Models on the Pareto chart
—
Filters
Chart settings
Intelligence Index vs. Cost per Intelligence Index Task
Artificial Analysis Intelligence Index · Weighted average cost (USD) per Artificial Analysis Intelligence Index task
Why these models?
Intelligence
Artificial Analysis Intelligence Index · Higher is better
Speed
Output tokens per second · Higher is better
Cost per Task
Weighted average cost (USD) per Intelligence Index task · Lower is better