MiniMax: MiniMax M3 Performance
Try MiniMax M3 on us. Starter credit covers your first requests, then one key reaches all 14 provider routes with automatic failover.
minimax/minimax-m3
Measured performance
Continuously sampled p50/p95 time-to-first-token (TTFT), effective throughput, and success rate for MiniMax: MiniMax M3. Effective throughput uses provider-reported output tokens over complete request time. Unsupported route and probe-configuration rows are separated from provider downtime, and no prompt or output content is stored.
| Provider | p50 TTFT | p95 TTFT | Effective throughput | Uptime | Config excluded | Availability samples |
|---|---|---|---|---|---|---|
| minimax | 1355 ms | 7645 ms | 33 tok/s n=1 | 100.00% | 1 provider_error |
18 |
| parasail | 759 ms | 759 ms | 370 tok/s n=1 | 100.00% | — | 1 |
| fireworks | 859 ms | 1055 ms | 314 tok/s n=1 | 100.00% | 1 provider_error |
2 |
| lyceum | 1177 ms | 1177 ms | — | 100.00% | — | 1 |
| novita | 1335 ms | 1335 ms | 136 tok/s n=1 | 100.00% | — | 1 |
| streamlake | 1713 ms | 1713 ms | 97 tok/s n=2 | 100.00% | — | 1 |
| telnyx | 2058 ms | 3145 ms | 233 tok/s n=2 | 100.00% | — | 2 |
| wandb | 2207 ms | 3020 ms | 300 tok/s n=2 | 100.00% | — | 8 |
| together | 3249 ms | 3249 ms | 150 tok/s n=2 | 100.00% | — | 1 |
| atlas-cloud | — | — | 93 tok/s n=1 | — | — | 0 |
| boundless | — | — | 134 tok/s n=1 | — | — | 0 |
| featherless | — | — | 31 tok/s n=1 | — | — | 0 |
| siliconflow | — | — | 118 tok/s n=2 | — | — | 0 |
Full provider & model leaderboard.
14 routes.
More routes give the auto router more room to fail over around provider 429 and 5xx responses.
Gateway overhead is measured separately.
Public status separates TLS/health overhead from full model latency so slow LLMs do not inflate the router metric.
Metadata rollups.
Status samples store latency, outcome, provider, model, route, cost, and region metadata only.
View public status or inspect provider routes.