Fastest LLM APIs (real-world speed ranking)

As of 2026-09-27, the top model is mimo-v2.6-pro-ultraspeed (408 tok/s). See the full ranking below — click a model for details or run your own test.

⚡ Fastest💰 Cheapest⭐ Best value
  1. 🥇
    mimo-v2.6-pro-ultraspeedapi.xiaomimimo.com
    TTFT 11.04s · peak 955 · 6 runs
    408 tok/s
  2. 🥈
    deepseek-flashapi.deepseek.com
    TTFT 0.62s · peak 571 · 14 runs
    322 tok/s
  3. 🥉
    deepseek-v4-flashapi.deepseek.com
    TTFT 2.83s · peak 258 · 5 runs
    284 tok/s
  4. 4
    deepseek-v4-proapi.deepseek.com
    TTFT 4.21s · peak 0 · 3 runs
    192 tok/s
  5. 5
    stealth/space-bunny-alphaopenrouter.ai
    TTFT 1.36s · peak 269 · 2 runs
    129 tok/s
  6. 6
    tencent/hy3-freeapi.orcarouter.ai
    TTFT 1.55s · peak 150 · 3 runs
    129 tok/s
  7. 7
    glm-5.3open.bigmodel.cn
    TTFT 2.29s · peak 227 · 13 runs
    114 tok/s
  8. 8
    glm-5.3-flashopen.bigmodel.cn
    TTFT 1.03s · peak 225 · 20 runs
    100 tok/s
  9. 9
    mimo-v2.6-flashapi.xiaomimimo.com
    TTFT 13.26s · peak 220 · 8 runs
    93 tok/s
  10. 10
    mimo-v2.6-proapi.xiaomimimo.com
    TTFT 2.31s · peak 239 · 10 runs
    54 tok/s
Full pricing →Speed leaderboard →Run your own test ▶
📋 Embed this ranking on your site

Copy the code below (live data, auto-updating):

<iframe src="https://www.tokrace.com/embed/best/fastest?lang=en" width="400" height="360" style="border:0;border-radius:12px" loading="lazy" title="Fastest LLM APIs (real-world speed ranking)"></iframe>

· Speed is the median of anonymous benchmarks; prices follow official pages and cross-currency uses a fixed USD rate. Reference only.