Gemma 4 31B vs MiniMax M3
API pricing, context, and independently measured performance, side by side. Gemma 4 31B starts 57% cheaper on input than MiniMax M3 at list prices.
Gemma 4 31B
Google · released Apr 2, 2026
Best input$0.10 / MTok
Best output$0.34 / MTok
Context262K
Providers17
Fastest measured1.4s° TTFT · 20 tok/s
Open weightsVisionToolsReasoningStructured outputPrompt caching
MiniMax M3
MiniMax · released May 31, 2026
Best input$0.23 / MTok
Best output$0.96 / MTok
Context1.05M
Providers14
Fastest measured0.52s° TTFT · 105 tok/s
Open weightsVisionToolsReasoningStructured outputPrompt caching
Cheapest providers, measuredUSD / MTOK
| MODEL | PROVIDER | INPUT | OUTPUT | TTFT | TPS | SMOKE |
|---|---|---|---|---|---|---|
| Gemma 4 31B |  CoreWeave | $0.10 | $0.34 | 3.1s° | 23° tok/s | 5/5 |
| Gemma 4 31B |  Venice | $0.12 | $0.36 | — | — | 5/5 |
| Gemma 4 31B |  Chutes | $0.12 | $0.37 | 4.2s° | 29° tok/s | 5/5 |
| MiniMax M3 | CoreWeave | $0.23 | $0.96 | — | — | 5/5 |
| MiniMax M3 |  GMICloud | $0.24 | $0.96 | 3.5s° | 150° tok/s | 5/5 |
| MiniMax M3 |  DeepInfra | $0.28 | $1.10 | 1.1s° | 23° tok/s | 5/5 |
TTFT/TPS: ModelIndex-measured medians (P50, trailing 21d) · chat10k profile (~10K in / 200 out) · ° = measured via OpenRouter routing (adds a hop) · SMOKE: of 5 programmatic capability probes passed · methodology v1.1 — full details on each model page