TOKRACE is a free, open-source LLM speed testing and AI model evaluation bench for developers and AI reviewers. It runs one prompt across multiple LLMs and compares TTFT, output TPS, peak speed and stability.

Is TOKRACE free to use?

You can try preset models within the free quota. Advanced users can add their own API keys; their provider may charge for requests. The project is open source.

Will my API key be exposed?

API keys are saved in browser localStorage by default and forwarded to your selected provider when making requests. Optional cloud sync uploads browser-encrypted backups, and publishing shares saves public snapshots. Avoid saving keys on shared devices. See the privacy policy.

Where does leaderboard data come from?

Speed and audience leaderboards use metrics and votes that users voluntarily share anonymously. They contain speed numbers only, not prompts or API keys.

Free model trials · Open-source benchmarks · Shareable evidence

A real-time model speed benchfor developers and AI reviewers

Run one prompt across multiple LLMs and compare TTFT · output TPS · peak speed · stability. Start with free model trials, then switch to your real task and model setup.

PREVIEWDeepSeek V4.1 Flash limited preview · API status and speed test

Run Sample ▶Professional Mode Templates Invite Rewards GitHub

Open Source

Write a 5,000-word essay on why speed itself is a capability● LIVE

Model A 🥇312 tok/s

Model B 🥈187 tok/s

Model C 🥉96 tok/s

0.68s

TTFT

102

Thinking TPS

312

Output TPS

387

Peak tok/s

↑ Demo animation · real data comes from your arena runs

Run a free sample

Try a free model evaluation with preset models, no registration or keys required.

Use your prompt

Test your real task, long-form input, code task or visual prompt against configured models.

Share evidence

Generate snapshots, voting pages and long images so readers can inspect the result.

Why test speed with TOKRACE

🏁

Concurrent tests with the same prompt

Send the same prompt to multiple models at the same time to reduce ordering, network and time-of-day bias.

⏱

Real-time TTFT / TPS comparison

Track TTFT, thinking TPS, output TPS, peak speed and token counts side by side.

📈

Zoomable speed curves

Inspect 2-second sliding-window tok/s curves, zoom into peak moments and export PNG evidence.

🔌

Free trials + bring your own key

New users can try preset models and run free evaluation samples without keys; advanced users can connect OpenAI-compatible or native Anthropic endpoints.

🔐

Keys saved locally by default

API keys are saved in your browser and forwarded when you make requests. Optional cloud backups are encrypted in your browser before upload.

📷

Built for publishing

Edit titles and notes, use screenshot mode, customize watermarks and copy metrics as Markdown.

🧪

Validate as you connect

Fetch provider model lists, test connectivity and configure private parameters per model.

🖼

Automatic SVG rendering

SVG code in model output is rendered as an image in a sandbox with scripts disabled.

NEW PROVIDER

Now supports OrcaRouter

Connect with PKCE or paste an sk-orca key to test the OrcaRouter Auto route and namespaced fixed models.

Integration guide →

Your keys stay yours

API keys are saved by default in your browser localStorage. Requests pass through a proxy to your selected provider. The proxy does not intentionally persist keys or request bodies. Publishing a share saves a public snapshot; optional cloud sync saves a browser-encrypted backup. Prefer no hosted proxy at all? The project is open source:git clone && npm run dev.

Run Sample ▶View source