Gemma 4 31B Garnet vs GLM 4.6
Gemma 4 31B Garnet is cheaper on output tokens, while Gemma 4 31B Garnet offers a larger context window. Choose Gemma 4 31B Garnet or GLM 4.6 based on the trade-off between cost, context, and the benchmarks that matter for your use case.
| Spec | Gemma 4 31B Garnet | GLM 4.6 |
|---|---|---|
| Provider | NanoGPT | Z.ai |
| Input / 1M tokens | $0.31 | $0.55 |
| Output / 1M tokens | $0.31 | $2.20 |
| Context window | 262K | 205K |
| Parameters | — | 357B |
| Open weights | No | Yes |
| Released | May 2026 | Sep 2025 |
| LiveBench score | — | 54.7 |
| LMArena (Chatbot Arena) Elo score | — | 1440 |
| FrontierMath score | — | 3.8 |
| LMArena WebDev Elo score | — | 1341 |
| SWE-bench Verified score | — | 68.2 |
FAQ
Is Gemma 4 31B Garnet or GLM 4.6 cheaper?
Gemma 4 31B Garnet is cheaper on output tokens ($0.31 vs $2.20 per 1M).
Which has the larger context window, Gemma 4 31B Garnet or GLM 4.6?
Gemma 4 31B Garnet has the larger context window (262K tokens).
Pricing is indicative — confirm with the provider before production use. Highlighted values indicate the better figure for that row.