Gemma 4 31B Split-Untied vs MiniMax M2.5
Gemma 4 31B Split-Untied is cheaper on output tokens, while Gemma 4 31B Split-Untied offers a larger context window. Choose Gemma 4 31B Split-Untied or MiniMax M2.5 based on the trade-off between cost, context, and the benchmarks that matter for your use case.
| Spec | Gemma 4 31B Split-Untied | MiniMax M2.5 |
|---|---|---|
| Provider | NanoGPT | NanoGPT |
| Input / 1M tokens | $0.10 | $0.30 |
| Output / 1M tokens | $0.30 | $1.20 |
| Context window | 262K | 205K |
| Parameters | — | — |
| Open weights | Yes | Yes |
| Released | Sep 2026 | Feb 2026 |
FAQ
Pricing is indicative — confirm with the provider before production use. Highlighted values indicate the better figure for that row.