API Providers Leaderboard
Compare the latency, speed, and pricing of top serverless inference providers for open-weights models.
| # | Provider | Models Available | Avg TTFT (s) | Avg Speed (t/s) | Pricing Level |
|---|---|---|---|---|---|
| 1 | Groq | 15+ | Medium | ||
| 2 | Fireworks AI | 85+ | Low | ||
| 3 | OctoAI | 25+ | Low | ||
| 4 | Together AI | 150+ | Low | ||
| 5 | Anyscale | 35+ | Medium | ||
| 6 | DeepInfra | 40+ | Lowest |