MiniMax M2.7: 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) | 99.59% | 98.82% | 98.82% | 5.54s | 63.7s | 4.0 tps | 197K | $0.300 / $1.20 | |
| Healthy | 100%◆ (best) | 99.71% | 99.73%◆ (best) | 99.73%◆ (best) | 527ms | 4.76s◆ (best) | 39 tps | 197K | $0.250 / $1.00 | |
| Healthy | 100%◆ (best) | 99.96% | 99.73% | 99.73% | 4.23s | 8.07s | 5.0 tps | 197K | $0.240 / $0.960◆ (best) | |
Groqgroq | Healthy | 100%◆ (best) | 99.90% | 99.54% | 99.54% | 322ms◆ (best) | 79.6s | 313 tps◆ (best) | 197K | $0.600 / $1.80 |
Maramara | Healthy | 100%◆ (best) | 99.51% | 99.02% | 99.02% | 1.83s | 18.4s | 81 tps | 197K | $0.240 / $0.960◆ (best) |
| Healthy | 99.70% | 99.20% | 98.29% | 98.29% | 2.08s | 7.19s | 29 tps | 205K | $0.300 / $1.20 | |
| Healthy | 99.44% | 92.70% | 97.28% | 97.28% | 1.54s | 9.13s | 41 tps | 205K | $0.600 / $2.40 | |
Fireworksfireworks | Healthy | 96.77% | 99.22% | 98.95% | 98.95% | 4.01s | 13.0s | 122 tps | 197K | $0.300 / $1.20 |
| No data | — | 42.14% | 81.08% | 81.08% | 2.34s | 21.3s | 79 tps | 197K | $0.380 / $1.70 | |
| No data | — | 99.25% | 98.23% | 98.23% | 3.92s | 8.71s | 6.0 tps | 205K | $0.270 / $1.08 | |
SambaNovasambanova | No data | — | 90.38% | 99.20% | 99.20% | 9.76s | 56.3s | 20 tps | 197K | $0.600 / $2.40 |
- AtlasCloud100%◆ (best)Healthy24h 99.59%
- 7d
- 98.82%
- 30d
- 98.82%
- p50
- 5.54s
- p99
- 63.7s
- Throughput
- 4.0 tps
- Context
- 197K
- $/Mtok
- $0.300 / $1.20
atlas-cloud/fp8fp8 - DeepInfra100%◆ (best)Healthy24h 99.71%
- 7d
- 99.73%◆ (best)
- 30d
- 99.73%◆ (best)
- p50
- 527ms
- p99
- 4.76s◆ (best)
- Throughput
- 39 tps
- Context
- 197K
- $/Mtok
- $0.250 / $1.00
deepinfra/fp8fp8 - GMICloud100%◆ (best)Healthy24h 99.96%
- 7d
- 99.73%
- 30d
- 99.73%
- p50
- 4.23s
- p99
- 8.07s
- Throughput
- 5.0 tps
- Context
- 197K
- $/Mtok
- $0.240 / $0.960◆ (best)
gmicloud/fp8fp8 - Groq100%◆ (best)Healthy24h 99.90%
- 7d
- 99.54%
- 30d
- 99.54%
- p50
- 322ms◆ (best)
- p99
- 79.6s
- Throughput
- 313 tps◆ (best)
- Context
- 197K
- $/Mtok
- $0.600 / $1.80
groq - Mara100%◆ (best)Healthy24h 99.51%
- 7d
- 99.02%
- 30d
- 99.02%
- p50
- 1.83s
- p99
- 18.4s
- Throughput
- 81 tps
- Context
- 197K
- $/Mtok
- $0.240 / $0.960◆ (best)
mara - Minimax99.70%Healthy24h 99.20%
- 7d
- 98.29%
- 30d
- 98.29%
- p50
- 2.08s
- p99
- 7.19s
- Throughput
- 29 tps
- Context
- 205K
- $/Mtok
- $0.300 / $1.20
minimax/fp8fp8 - Minimax99.44%Healthy24h 92.70%
- 7d
- 97.28%
- 30d
- 97.28%
- p50
- 1.54s
- p99
- 9.13s
- Throughput
- 41 tps
- Context
- 205K
- $/Mtok
- $0.600 / $2.40
minimax/highspeedfp8 - Fireworks96.77%Healthy24h 99.22%
- 7d
- 98.95%
- 30d
- 98.95%
- p50
- 4.01s
- p99
- 13.0s
- Throughput
- 122 tps
- Context
- 197K
- $/Mtok
- $0.300 / $1.20
fireworks - No data24h 42.14%
- 7d
- 81.08%
- 30d
- 81.08%
- p50
- 2.34s
- p99
- 21.3s
- Throughput
- 79 tps
- Context
- 197K
- $/Mtok
- $0.380 / $1.70
deepinfra/turbofp8 - No data24h 99.25%
- 7d
- 98.23%
- 30d
- 98.23%
- p50
- 3.92s
- p99
- 8.71s
- Throughput
- 6.0 tps
- Context
- 205K
- $/Mtok
- $0.270 / $1.08
novita/fp8fp8 - No data24h 90.38%
- 7d
- 99.20%
- 30d
- 99.20%
- p50
- 9.76s
- p99
- 56.3s
- Throughput
- 20 tps
- Context
- 197K
- $/Mtok
- $0.600 / $2.40
sambanova
Measured 4m 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.
Throughput
Median tokens per second.