Fastest LLM APIs
LLM APIs ranked by measured output throughput in tokens per second. Speed-focused providers like Groq, Cerebras, and SambaNova lead here; throughput is reported where we have data.
Ranking 2 options · updated July 4, 2026
| # | Model | Provider | Tokens/sec |
|---|---|---|---|
| 1 | Kimi K2.7 Code open | Nebius | 231.26 |
| 2 | Kimi K2.6 open | Nebius | 60 |