MiniMax: MiniMax M3 Performance
Compare measured TTFT, throughput, uptime, and route health for MiniMax: MiniMax M3 across TrustedRouter providers using metadata-only production probes.
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 |
|---|---|---|---|---|---|---|
| parasail | 992 ms | 3061 ms | 96 tok/s n=2 | 66.67% | — | 3 |
| together | 995 ms | 995 ms | 90 tok/s n=2 | 100.00% | — | 16 |
| telnyx | 1224 ms | 2580 ms | 114 tok/s n=2 | 100.00% | — | 4 |
| fireworks | 1391 ms | 1391 ms | 105 tok/s n=2 | 100.00% | — | 1 |
| nvidia-nim | 1561 ms | 1561 ms | 35 tok/s n=2 | 100.00% | — | 1 |
| minimax | 3100 ms | 10158 ms | 52 tok/s n=2 | 100.00% | — | 7 |
| morph | 3293 ms | 5446 ms | 35 tok/s n=1 | 66.67% | — | 3 |
| deepinfra | — | — | 19 tok/s n=2 | 100.00% | — | 18 |
| featherless | — | — | 118 tok/s n=2 | — | 3 probe_config_error |
0 |
| novita | — | — | 57 tok/s n=2 | — | — | 0 |
| siliconflow | — | — | 73 tok/s n=2 | — | — | 0 |
| wandb | — | — | 83 tok/s n=2 | — | 2 probe_config_error |
0 |
Full provider & model leaderboard.
13 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.