MiniMax M2.5: providers compared
Every endpoint serving this model, side by side. ◆ marks the best value in each column — per model, never fleet-wide.
Current state
| State | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Healthy | 100%◆ (best) | 99.79% | 99.86% | 99.86% | 1.75s | 10.0s | 44 tps | 197K | $0.295 / $1.20 | |
DigitalOceandigitalocean | Healthy | 100%◆ (best) | 99.91% | 99.87% | 99.87% | 591ms | 3.76s | 62 tps | 66K | $0.300 / $1.20 |
| Healthy | 100%◆ (best) | 99.84% | 99.83% | 99.83% | 1.34s | 3.81s | 54 tps | 205K | $0.300 / $1.20 | |
| Healthy | 100%◆ (best) | 100% | 100%◆ (best) | 100%◆ (best) | 1.37s | 3.02s◆ (best) | 45 tps | 197K | $0.300 / $1.20 | |
StreamLakestreamlake | Healthy | 99.29% | 99.88% | 99.93% | 99.93% | 777ms | 5.03s | 59 tps | 200K | $0.270 / $1.08 |
Friendlifriendli | No data | — | 99.67% | 99.95% | 99.95% | 237ms◆ (best) | 3.24s | 107 tps◆ (best) | 197K | $0.300 / $1.20 |
| No data | — | — | 99.99% | 99.99% | — | — | — | 197K | $0.220 / $0.900◆ (best) | |
| No data | — | 99.82% | 99.81% | 99.81% | 1.86s | 4.60s | 61 tps | 205K | $0.300 / $1.20 | |
| No data | — | 99.43% | 100%◆ (best) | 100%◆ (best) | 2.31s | 7.72s | 39 tps | 205K | $0.600 / $2.40 | |
Venicevenice | Degraded | — | 99.45% | 99.81% | 99.81% | 1.62s | 22.2s | 34 tps | 198K | $0.270 / $0.950 |
- AtlasCloud100%◆ (best)Healthy24h 99.79%
- 7d
- 99.86%
- 30d
- 99.86%
- p50
- 1.75s
- p99
- 10.0s
- Throughput
- 44 tps
- Context
- 197K
- $/Mtok
- $0.295 / $1.20
atlas-cloud/fp8fp8 - DigitalOcean100%◆ (best)Healthy24h 99.91%
- 7d
- 99.87%
- 30d
- 99.87%
- p50
- 591ms
- p99
- 3.76s
- Throughput
- 62 tps
- Context
- 66K
- $/Mtok
- $0.300 / $1.20
digitalocean - Novita100%◆ (best)Healthy24h 99.84%
- 7d
- 99.83%
- 30d
- 99.83%
- p50
- 1.34s
- p99
- 3.81s
- Throughput
- 54 tps
- Context
- 205K
- $/Mtok
- $0.300 / $1.20
novita/fp8fp8 - SiliconFlow100%◆ (best)Healthy24h 100%
- 7d
- 100%◆ (best)
- 30d
- 100%◆ (best)
- p50
- 1.37s
- p99
- 3.02s◆ (best)
- Throughput
- 45 tps
- Context
- 197K
- $/Mtok
- $0.300 / $1.20
siliconflow/fp8fp8 - StreamLake99.29%Healthy24h 99.88%
- 7d
- 99.93%
- 30d
- 99.93%
- p50
- 777ms
- p99
- 5.03s
- Throughput
- 59 tps
- Context
- 200K
- $/Mtok
- $0.270 / $1.08
streamlake - No data24h 99.67%
- 7d
- 99.95%
- 30d
- 99.95%
- p50
- 237ms◆ (best)
- p99
- 3.24s
- Throughput
- 107 tps◆ (best)
- Context
- 197K
- $/Mtok
- $0.300 / $1.20
friendli - No data24h —
- 7d
- 99.99%
- 30d
- 99.99%
- p50
- —
- p99
- —
- Throughput
- —
- Context
- 197K
- $/Mtok
- $0.220 / $0.900◆ (best)
inceptron/fp8fp8 - No data24h 99.82%
- 7d
- 99.81%
- 30d
- 99.81%
- p50
- 1.86s
- p99
- 4.60s
- Throughput
- 61 tps
- Context
- 205K
- $/Mtok
- $0.300 / $1.20
minimax/fp8fp8 - No data24h 99.43%
- 7d
- 100%◆ (best)
- 30d
- 100%◆ (best)
- p50
- 2.31s
- p99
- 7.72s
- Throughput
- 39 tps
- Context
- 205K
- $/Mtok
- $0.600 / $2.40
minimax/highspeedfp8 - Degraded24h 99.45%
- 7d
- 99.81%
- 30d
- 99.81%
- p50
- 1.62s
- p99
- 22.2s
- Throughput
- 34 tps
- Context
- 198K
- $/Mtok
- $0.270 / $0.950
venice
Measured 3m ago. Latency and throughput are OpenRouter's rolling 30-minute windows. ◆ marks the best value in each column among these endpoints.
Price vs latency
One dot per endpoint at the latest round · further left = faster, lower = cheaper · hover a dot for its tag
History
Availability
Axis starts at 90% so small differences stay visible; it expands to 0 when an endpoint falls below.
Time to first token
Logarithmic axis — the fleet spans two orders of magnitude. P50 is OpenRouter's rolling 30-minute window, averaged over the hour.
Throughput
Median tokens per second.