MiniMax: MiniMax M2.5
minimax/minimax-m2.5
Cheapest provider
$0.15 / 1M
AkashML
Fastest provider (p95)
—
No throughput data yet — populated as traffic accumulates
Intelligence (estimated)
72.0
Family + generation + popularity + price tier
Per-provider performance
Latency / throughput / uptime measured across providers over the last 30 minutes of live traffic. Atlas’s router weighs these per call (with the eval-gate signal) when picking a variant for Standard and Batch tiers.
| Provider | Quant | Input $/1M | Output $/1M | Latency p50 / p95 | Throughput p50 / p95 | Uptime 30m | Success |
|---|---|---|---|---|---|---|---|
| AkashML | q8· fp8 | $0.1500 | $1.1500 | — | — | 99.19% | — |
| Chutes | q8· fp8 | $0.1500 | $1.2000 | — | — | 100.00% | — |
| DeepInfra | q8· fp8 | $0.1500 | $1.1500 | — | — | 99.99% | — |
| Phala | undisclosed | $0.2000 | $1.3800 | — | — | 100.00% | — |
| Inceptron | q8· fp8 | $0.2400 | $0.9000 | — | — | 98.18% | — |
| Baidu | q8· fp8 | $0.2700 | $1.0800 | — | — | 100.00% | — |
| AtlasCloud | q8· fp8 | $0.2950 | $1.2000 | — | — | 98.13% | — |
| Friendli | undisclosed | $0.3000 | $1.2000 | — | — | — | — |
| Mara | undisclosed | $0.3000 | $1.2000 | — | — | 100.00% | — |
| Minimax | q8· fp8 | $0.3000 | $1.2000 | — | — | 97.33% | — |
| Novita | q8· fp8 | $0.3000 | $1.2000 | — | — | — | — |
| Parasail | q8· fp8 | $0.3000 | $1.2000 | — | — | 100.00% | — |
| SiliconFlow | q8· fp8 | $0.3000 | $1.2000 | — | — | — | — |
| StreamLake | undisclosed | $0.3000 | $1.2000 | — | — | — | — |
| WandB | q8· fp8 | $0.3000 | $1.2000 | — | — | — | — |
| Venice | undisclosed | $0.3400 | $1.1900 | — | — | — | — |
| Minimax | q8· fp8 | $0.6000 | $2.4000 | — | — | — | — |
“—” means live telemetry hasn’t accumulated enough recent traffic for that endpoint. “undisclosed” means the provider serves the model but doesn’t expose the quantization label (typically running fp8 / int8 internally).
Intelligence estimate
No public benchmark numbers indexed for this model yet, so the leaderboard score is derived from the catalogue data we sync: model family standing, generation, and live usage rank — applied identically to open- and closed-weight models. The heuristic ceiling sits below confirmed-benchmark frontier models so curated rankings stay clearly on top.
- ·minimax family base
Want a curated score? File a benchmark report and we’ll add it to the next sync.
How Atlas routes MiniMax: MiniMax M2.5
- Realtime tier — direct passthrough at the upstream’s native precision. Best for hard latency / quality guarantees.
- Standard tier — Atlas picks the cheapest provider variant whose quantization has stayed green on your operation’s eval gates. For MiniMax: MiniMax M2.5 that’s currently AkashML at $0.15/1M.
- Batch tier — async, biggest discount. Roadmapped to use provider batch APIs (OpenAI / Anthropic) where available and queued spot capacity for open-weight workloads.