Observed arrival · 2026-09-10
LLM API Check: Browser Benchmarks for Real Endpoints
An open-source browser tool for comparing LLM latency, throughput, capabilities, and message-format compatibility across providers and OpenAI-compatible endpoints.
Field notes
The tool sends benchmark requests from the browser to the configured provider rather than routing them through a site backend. Its probes go beyond headline speed: they include randomized prompts to reduce cache distortion, warm and cold cache comparisons, synthetic-image inputs, local JSON Schema validation, and message shapes commonly produced by real agents. The homepage says keys, model lists, and results remain in browser storage, while CORS restrictions can prevent direct calls to some providers.
Observed signals
Read the marks
Editorial observations of this landing page, not a rating.
One card from the complete issue