Gemma 4 31B
Largest Gemma 4 instruction model for open, self-hosted chat and reasoning
Model facts
- Context window
- 262K tokens
- Maximum output
- 33K tokens
- Cheapest paid input
- $0.102 per 1M tokens
- Cheapest paid output
- $0.297 per 1M tokens
- Open weights
- yes
- Providers
- 42
Capabilities and modalities
reasoning, tool calling, structured output, attachments, temperature control, open weights, text, image, audio, video, pdf.
Local hardware estimate
At 4K context: Q4_K_M 20.1 GiB, Q8_0 34.7 GiB, F16 62.0 GiB working memory.
- Parameters
- 31.3 billion
- Architecture
- gemma4
- Model license
- apache-2.0
Planning estimate adapted from the MIT-licensed whichllm estimator using metadata from Hugging Face; not a vendor minimum requirement.
Artificial Analysis benchmarks
Compare this model on the full LLM benchmark leaderboard.
| Configuration | AAI | Coding | Math | Output tokens/s |
|---|---|---|---|---|
| Reasoning | 29.7 | 43.4 | not measured | 35.2 |
| Non-reasoning | 22.3 | 33.2 | not measured | 73.0 |
Observed price history
- 2026-08-24: $0.102 input / $0.297 output per 1M tokens
API providers
- Kenari — model id gemma-4-31b-it; free input / free output per 1M tokens; provider documentation
- Nvidia — model id google/gemma-4-31b-it; free input / free output per 1M tokens; provider documentation
- OpenRouter (free) — model id google/gemma-4-31b-it:free; free input / free output per 1M tokens; provider documentation
- Poe — model id google/gemma-4-31b; free input / free output per 1M tokens; provider documentation
- QVAC — model id gemma4-31b; free input / free output per 1M tokens; provider documentation
- Requesty — model id gemma-4-31b-it; free input / free output per 1M tokens; provider documentation
- UnoRouter — model id gemma-4-31b-it:free; free input / free output per 1M tokens; provider documentation
- DevPass (LLM Gateway) — model id gemma-4-31b-it; $0.102 input / $0.297 output per 1M tokens; provider documentation
- CrofAI — model id gemma-4-31b-it; $0.1 input / $0.3 output per 1M tokens; provider documentation
- OpenRouter — model id google/gemma-4-31b-it; $0.09 input / $0.34 output per 1M tokens; provider documentation
- Kilo Gateway — model id google/gemma-4-31b-it; $0.09 input / $0.34 output per 1M tokens; provider documentation
- Weights & Biases — model id google/gemma-4-31B-it; $0.1 input / $0.34 output per 1M tokens; provider documentation
- Lilac — model id google/gemma-4-31b-it; $0.11 input / $0.35 output per 1M tokens; provider documentation
- Venice AI — model id google-gemma-4-31b-it; $0.12 input / $0.36 output per 1M tokens; provider documentation
- Deep Infra — model id google/gemma-4-31B-it; $0.13 input / $0.38 output per 1M tokens; provider documentation
- FastRouter — model id google/gemma-4-31b-it; $0.13 input / $0.38 output per 1M tokens; provider documentation
- OrcaRouter — model id google/gemma-4-31b-it; $0.13 input / $0.38 output per 1M tokens; provider documentation
- NEAR AI Cloud — model id google/gemma-4-31B-it; $0.13 input / $0.4 output per 1M tokens; provider documentation
- SiliconFlow — model id google/gemma-4-31B-it; $0.13 input / $0.4 output per 1M tokens; provider documentation
- Abacus — model id google/gemma-4-31b-it; $0.14 input / $0.4 output per 1M tokens; provider documentation
- Crusoe — model id google/gemma-4-31b-it; $0.14 input / $0.4 output per 1M tokens; provider documentation
- Friendli — model id google/gemma-4-31B-it; $0.14 input / $0.4 output per 1M tokens; provider documentation
- Hugging Face — model id google/gemma-4-31B-it; $0.14 input / $0.4 output per 1M tokens; provider documentation
- Merge Gateway — model id google/gemma-4-31b-it; $0.14 input / $0.4 output per 1M tokens; provider documentation
- NovitaAI — model id google/gemma-4-31b-it; $0.14 input / $0.4 output per 1M tokens; provider documentation
- Vercel AI Gateway — model id google/gemma-4-31b-it; $0.14 input / $0.4 output per 1M tokens; provider documentation
- NanoGPT — model id google/gemma-4-31b-it; $0.1 input / $0.45 output per 1M tokens; provider documentation
- Neuralwatt — model id gemma-4-31b; $0.144 input / $0.42 output per 1M tokens; provider documentation
- Cortecs — model id gemma-4-31b-it; $0.223 input / $0.39 output per 1M tokens; provider documentation
- ai& — model id google/gemma-4-31b-it; $0.2 input / $0.5 output per 1M tokens; provider documentation
- Infomaniak — model id google/gemma-4-31B-it; $0.25 input / $0.5 output per 1M tokens; provider documentation
- Berget.AI — model id google/gemma-4-31B-it; $0.275 input / $0.55 output per 1M tokens; provider documentation
- Pioneer — model id google/gemma-4-31B-it; $0.5 input / $0.5 output per 1M tokens; provider documentation
- Together AI — model id google/gemma-4-31B-it; $0.39 input / $0.97 output per 1M tokens; provider documentation
- OpenRouter (batch) — model id google/gemma-4-31b-it:batch; $0.39 input / $0.97 output per 1M tokens; provider documentation
- NanoGPT — model id TEE/gemma4-31b; $0.4 input / $1 output per 1M tokens; provider documentation
- Tinfoil — model id gemma4-31b; $0.4 input / $1 output per 1M tokens; provider documentation
- Cerebras — model id gemma-4-31b; $0.99 input / $1.49 output per 1M tokens; provider documentation
- Regolo AI — model id gemma4-31b; $0.46 input / $2.42 output per 1M tokens; provider documentation
- Arena — model id pteronura; not published input / not published output per 1M tokens; provider documentation
- Google — model id gemma-4-31b-it; not published input / not published output per 1M tokens; provider documentation
- Ollama Cloud — model id gemma4:31b; not published input / not published output per 1M tokens; provider documentation