MiMo-V2.5: 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) | 97.84% | 98.61% | 98.61% | 2.40s | 19.3s | 38 tps | 1.0M | $0.168 / $0.336 | |
| Healthy | 100%◆ (best) | 99.64% | 98.35% | 98.35% | 339ms◆ (best) | 4.93s | 72 tps◆ (best) | 1.0M | $0.140 / $0.280 | |
| Healthy | 99.87% | 99.67% | 99.05%◆ (best) | 99.05%◆ (best) | 2.82s | 31.0s | 34 tps | 1.0M | $0.140 / $0.280 | |
| Healthy | 98.37% | 98.72% | 95.58% | 95.58% | 477ms | 1.98s◆ (best) | 31 tps | 262K | $0.400 / $2.00 | |
| Degraded | 91.40% | 86.73% | 88.36% | 88.36% | 4.01s | 8.11s | 2.0 tps | 1.1M | $0.119 / $0.238◆ (best) |
- Novita100%◆ (best)Healthy24h 97.84%
- 7d
- 98.61%
- 30d
- 98.61%
- p50
- 2.40s
- p99
- 19.3s
- Throughput
- 38 tps
- Context
- 1.0M
- $/Mtok
- $0.168 / $0.336
novita/fp8fp8 - Parasail100%◆ (best)Healthy24h 99.64%
- 7d
- 98.35%
- 30d
- 98.35%
- p50
- 339ms◆ (best)
- p99
- 4.93s
- Throughput
- 72 tps◆ (best)
- Context
- 1.0M
- $/Mtok
- $0.140 / $0.280
parasail/fp8fp8 - Xiaomi99.87%Healthy24h 99.67%
- 7d
- 99.05%◆ (best)
- 30d
- 99.05%◆ (best)
- p50
- 2.82s
- p99
- 31.0s
- Throughput
- 34 tps
- Context
- 1.0M
- $/Mtok
- $0.140 / $0.280
xiaomi/fp8fp8 - DeepInfra98.37%Healthy24h 98.72%
- 7d
- 95.58%
- 30d
- 95.58%
- p50
- 477ms
- p99
- 1.98s◆ (best)
- Throughput
- 31 tps
- Context
- 262K
- $/Mtok
- $0.400 / $2.00
deepinfra/bf16bf16 - GMICloud91.40%Degraded24h 86.73%
- 7d
- 88.36%
- 30d
- 88.36%
- p50
- 4.01s
- p99
- 8.11s
- Throughput
- 2.0 tps
- Context
- 1.1M
- $/Mtok
- $0.119 / $0.238◆ (best)
gmicloud/fp8fp8
Measured 2m 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.