MiniMax-M3
Intelligence
29.61
#112 of 520 models
Output speed
168.7 tok/s
Measured generation rate
Cost / index task
$0.508
Measured evaluation workload
End-to-end latency
16ms
First answer 13ms
01 / Position
A model in the landscape
Compare this model with peers across intelligence, speed, cost, and individual evaluations.
02 / Evidence
Benchmark results
29 distinct evaluations grouped by topic. Explore cross-model charts or switch to the full results list.
Domain capability
03 / Operation
Speed, tokens & price
Prices are per million tokens; latency and task costs describe the captured evaluation rather than a universal provider guarantee.
Token pricing
- Input / 1M
- $0.300
- Output / 1M
- $1.20
- Cache read / 1M
- $0.060
- Cache write / 1M
- $0.375
- Cache-hit discount
- 80%
Response & usage
- Output speed
- 168.7 tok/s
- First chunk
- 1ms
- First answer
- 13ms
- Output / task
- 48.2k
- Reasoning / task
- 21.8k
Evaluation workload
- Canonical input
- 5.27B
- Canonical output
- 117.06M
- Answer / task
- 26.4k
- Index time / task
- 4.4m
- Run cost
- $537.978
By prompt type
| Prompt | Output speed | First chunk | First answer | End to end |
|---|---|---|---|---|
| Medium | 120.4 tok/s | 1ms | 18ms | 22ms |
| Long | 168.7 tok/s | 1ms | 13ms | 16ms |
| Hundred K | 134.8 tok/s | 4ms | 18ms | 22ms |
| Medium Parallel | 122.2 tok/s | 1ms | 17ms | 21ms |
04 / Access
License & availability
Recorded license
MINIMAX COMMUNITY LICENSE
The publisher-linked license text defines permitted use and redistribution. Review it before using the model.
- Weight access
- Open weights
- Classification
- Commercial License
- Reasoning
- Yes
Routing and providers
- OpenRouter ID
- minimax/minimax-m3
- Tokenizer
- Other
- Instruct type
- —
- Captured endpoints
- 13
| Provider | Quantization | Context | tok/s |
|---|---|---|---|
| CoreWeave | fp4 | 262.1k | — |
| GMICloud | fp8 | 1.05M | — |
| DeepInfra | fp8 | 524.3k | — |
| StreamLake | fp8 | 1M | — |
| Venice | fp8 | 524.3k | — |
| Together | unknown | 524.3k | — |
| Parasail | fp8 | 1.05M | — |
| AtlasCloud | fp8 | 524.3k | — |
| Novita | fp8 | 1M | — |
| Minimax | fp8 | 524.3k | — |
| Mara | unknown | 1.05M | — |
| SambaNova | unknown | 1.05M | — |
| ModelRun | fp4 | 1.05M | — |
05 / Audit
Underlying fields
Inspect the captured model and routing records. Artificial Analysis destination links are omitted.