Gemma 4 31B
Launched 2026-04-02 · Knowledge up to 2025-01 · Listed on OpenRouter 2026-04-02
Prices updated . Benchmarks updated .
Gemma 4 31B Instruct is Google DeepMind's 30.7B dense multimodal model supporting text and image input with text output. Features a 256K token context window, configurable thinking/reasoning mode, native function...
You send / 1M
$0.200
tokens in
It replies / 1M
$0.400
tokens out
Context
262K
tokens it can read at once
Typical request
$0.0005
1,000 words in, 400 out
Benchmarks
| Benchmark | Score | Kind | Date | Source |
|---|---|---|---|---|
| Epoch Capabilities Index | 143.1 | Independent | 2026-04-02 | Epoch AI |
| GPQA Diamond | 75.8% | Independent | 2026-08-06 | Epoch AI |
| OTIS Mock AIME | 73.3% | Independent | 2026-08-06 | Epoch AI |
| SimpleQA Verified | 10.4% | Independent | 2026-08-27 | Epoch AI |
| CritPt | 1.4% | Independent | 2026-04-02 | CritPt, republished by Epoch AI |
| Arena Elo (text) | 1,443 | Independent | 2026-10-28 | LMArena |
| Arena Elo (webdev) | 1,369 | Independent | 2026-10-28 | LMArena |
| Arena Elo (vision) | 1,276 | Independent | 2026-10-19 | LMArena |
Rankings on this site use the Epoch Capabilities Index, an independent composite. Lab-reported rows are the lab's own figures and are kept separate.
Data privacy
No training or retention: Yes. At least one provider serves it with zero data retention.
| Provider | You send / 1M | It replies / 1M | Uptime | Privacy policy |
|---|---|---|---|---|
| PrimeIntellect · fp8 | $0.051 | $0.085 | 99% | Read |
| Modal | $0.340 | $0.100 | 92% | Read |
| DeepInfra · fp8 | $0.160 | $0.320 | 98% | Read |
| CoreWeave · fp4 | $0.100 | $0.340 | 100% | Read |
| Venice · fp4 | $0.120 | $0.360 | 100% | Read |
| NextBit · fp4 | $0.150 | $0.400 | 100% | Read |
| Crusoe · bf16 | $0.140 | $0.400 | 100% | Read |
| DeepInfra · fp8 | $0.200 | $0.400 | 100% | Read |
| Parasail · fp8 | $0.150 | $0.400 | 100% | Read |
| Novita · bf16 | $0.140 | $0.400 | 98% | Read |
| DeepInfra · fp4 | $0.135 | $0.510 | 100% | Read |
| DeepInfra · fp8 | $0.300 | $0.600 | 98% | Read |
| ModelRun · fp4 | $0.750 | $1.00 | 100% | Read |
| SiliconFlow · fp8 | $0.750 | $1.00 | 94% | Read |
| SambaNova | $0.380 | $1.15 | 85% | Read |
Based on OpenRouter and Vercel AI Gateway data. Confirm terms with the provider before sending sensitive data.
What it can do
- Reads: image, text, video
- Writes: text
- Weights: Open weights, restricted
- Reasoning: Yes
- Tool calling: Yes
- Longest reply: 33K tokens
Compare with
Appears in
Estimate your monthly cost with Gemma 4 31B
Sources and licenses
- OpenRouter: Prices, context windows, modalities and the model list come from the OpenRouter public API. Data: OpenRouter.
- Vercel AI Gateway: The no-training flag comes from the Vercel AI Gateway model list.
- models.dev: Lab launch dates, open-weights flags and knowledge cutoffs come from models.dev. Copyright (c) 2025 models.dev, MIT License.
- Epoch AI: Independent benchmark scores and the Epoch Capabilities Index (ECI) come from Epoch AI's Benchmarking Hub, used under CC BY 4.0. Epoch AI, Capabilities & benchmarking, epoch.ai/benchmarks. Scores Epoch marks as external keep their original license, and the source is named on each score.
- LMArena: Arena Elo (text, webdev and vision) comes from the LMArena leaderboard dataset (lmarena-ai/leaderboard-dataset), CC BY 4.0, leaderboard date 2026-10-28.
- Lab-reported scores, when an editor has added them, are the lab's own numbers. Each one links to the source and is labeled lab-reported. They are not used to rank the lists.
- What a score measures: the benchmark guide.
