MiniMax M3
minimax/minimax-m3
Compare with MiniMax M2.7, MiniMax M2.5, MiniMax M3 (batch)
CompareMiniMax M3
| State | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Healthy | 100% | 99.70% | 99.52% | 99.52% | 980ms | 4.79s | 84 tps | 524K | $0.300 / $1.20 | |
| Healthy | 100% | 99.54% | 99.63% | 99.63% | 1.39s | 16.1s | 57 tps | 524K | $0.300 / $1.20 | |
Togethertogether | Healthy | 100% | 99.78% | 99.78% | 99.78% | 887ms | 10.3s | 73 tps | 524K | $0.300 / $1.20 |
| Healthy | 100% | 99.81% | 99.87% | 99.87% | 1.14s | 6.62s | 105 tps | 524K | $0.300 / $1.20 | |
| Healthy | 99.69% | 99.63% | 99.66% | 99.66% | 2.28s | 15.4s | 42 tps | 1.0M | $0.300 / $1.20 | |
| Healthy | 99.56% | 99.77% | 99.76% | 99.76% | 1.93s | 21.8s | 82 tps | 1.0M | $0.750 / $3.00 | |
| Healthy | 99.42% | 99.68% | 97.24% | 97.24% | 566ms | 18.0s | 92 tps | 262K | $0.230 / $0.960 | |
| Healthy | 99.27% | 96.69% | 97.49% | 97.49% | 1.50s | 21.0s | 21 tps | 256K | $0.255 / $1.02 | |
SambaNovasambanova | Healthy | 98.82% | 97.48% | 98.38% | 98.38% | 2.36s | 23.3s | 122 tps | 1.0M | $0.600 / $2.40 |
| Healthy | 96.99% | 98.41% | 98.38% | 98.38% | 943ms | 24.4s | 27 tps | 524K | $0.280 / $1.10 | |
| Healthy | 96.72% | 99.49% | 99.33% | 99.33% | 3.52s | 12.4s | 40 tps | 1.0M | $0.240 / $0.960 | |
| Healthy | 95.92% | 99.02% | 98.46% | 98.46% | 1.66s | 7.55s | 62 tps | 1.0M | $0.300 / $1.20 | |
| Down | 88.41% | 98.35% | 97.91% | 97.91% | 947ms | 12.7s | 50 tps | 1.0M | $0.300 / $1.20 |
- AtlasCloud100%Healthy24h 99.70%
- 7d
- 99.52%
- 30d
- 99.52%
- p50
- 980ms
- p99
- 4.79s
- Throughput
- 84 tps
- Context
- 524K
- $/Mtok
- $0.300 / $1.20
atlas-cloud/fp8fp8 - Minimax100%Healthy24h 99.54%
- 7d
- 99.63%
- 30d
- 99.63%
- p50
- 1.39s
- p99
- 16.1s
- Throughput
- 57 tps
- Context
- 524K
- $/Mtok
- $0.300 / $1.20
minimax/fp8fp8 - Together100%Healthy24h 99.78%
- 7d
- 99.78%
- 30d
- 99.78%
- p50
- 887ms
- p99
- 10.3s
- Throughput
- 73 tps
- Context
- 524K
- $/Mtok
- $0.300 / $1.20
together - Venice100%Healthy24h 99.81%
- 7d
- 99.87%
- 30d
- 99.87%
- p50
- 1.14s
- p99
- 6.62s
- Throughput
- 105 tps
- Context
- 524K
- $/Mtok
- $0.300 / $1.20
venice/fp8fp8 - Novita99.69%Healthy24h 99.63%
- 7d
- 99.66%
- 30d
- 99.66%
- p50
- 2.28s
- p99
- 15.4s
- Throughput
- 42 tps
- Context
- 1.0M
- $/Mtok
- $0.300 / $1.20
novita/fp8fp8 - ModelRun99.56%Healthy24h 99.77%
- 7d
- 99.76%
- 30d
- 99.76%
- p50
- 1.93s
- p99
- 21.8s
- Throughput
- 82 tps
- Context
- 1.0M
- $/Mtok
- $0.750 / $3.00
modelrun/fp4fp4 - CoreWeave99.42%Healthy24h 99.68%
- 7d
- 97.24%
- 30d
- 97.24%
- p50
- 566ms
- p99
- 18.0s
- Throughput
- 92 tps
- Context
- 262K
- $/Mtok
- $0.230 / $0.960
coreweave/fp4fp4 - Morph99.27%Healthy24h 96.69%
- 7d
- 97.49%
- 30d
- 97.49%
- p50
- 1.50s
- p99
- 21.0s
- Throughput
- 21 tps
- Context
- 256K
- $/Mtok
- $0.255 / $1.02
morph/fp4fp4 - SambaNova98.82%Healthy24h 97.48%
- 7d
- 98.38%
- 30d
- 98.38%
- p50
- 2.36s
- p99
- 23.3s
- Throughput
- 122 tps
- Context
- 1.0M
- $/Mtok
- $0.600 / $2.40
sambanova - DeepInfra96.99%Healthy24h 98.41%
- 7d
- 98.38%
- 30d
- 98.38%
- p50
- 943ms
- p99
- 24.4s
- Throughput
- 27 tps
- Context
- 524K
- $/Mtok
- $0.280 / $1.10
deepinfra/fp8fp8 - GMICloud96.72%Healthy24h 99.49%
- 7d
- 99.33%
- 30d
- 99.33%
- p50
- 3.52s
- p99
- 12.4s
- Throughput
- 40 tps
- Context
- 1.0M
- $/Mtok
- $0.240 / $0.960
gmicloud/fp8fp8 - StreamLake95.92%Healthy24h 99.02%
- 7d
- 98.46%
- 30d
- 98.46%
- p50
- 1.66s
- p99
- 7.55s
- Throughput
- 62 tps
- Context
- 1.0M
- $/Mtok
- $0.300 / $1.20
streamlake/fp8fp8 - Parasail88.41%Down24h 98.35%
- 7d
- 97.91%
- 30d
- 97.91%
- p50
- 947ms
- p99
- 12.7s
- Throughput
- 50 tps
- Context
- 1.0M
- $/Mtok
- $0.300 / $1.20
parasail/fp8fp8
Measured 2m ago. Latency and throughput are OpenRouter's rolling 30-minute windows.
Price changes
- MiniMax M3Morphmorph/fp4$0.300 / $1.20 → $0.255 / $1.022026-08-21 01:15 UTC
- MiniMax M3SambaNovasambanovafree / free → $0.600 / $2.402026-08-20 21:35 UTC
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.