Gemma 4 31B (Reasoning)
Intelligence
39.2
#90/523
Coding
38.7
#62/429
Math
—
Speed
35 tok/s
TTFT: 822.00s
Pricing
—
Gemma 4 31B (Reasoning) is Google’s AI model designed for advanced reasoning tasks. It processes at 35.377 tokens per second and is priced at $0 per million tokens, targeting professional users.
When to Use Gemma 4 31B (Reasoning)
✓ Best For
- Complex problem-solving tasks.
- Data analysis and interpretation.
- Software development assistance.
✗ Not Ideal For
- Basic conversational applications.
- Tasks requiring high-speed processing.
How Gemma 4 31B (Reasoning) Compares
Intelligence Index · Higher is better
MiniMaxDeepSeekGoogleMistralOpenAI
Benchmark Profile
Coding Index
OpenAIDeepSeekGoogleAnthropicKimi
Output Speed · tok/s
MistralAnthropicGoogleKimi
Intelligence · Coding · Math
Intelligence
Coding
Math
All Benchmark Scores (9)
| Benchmark | Score | |
|---|---|---|
| Intelligence Index | 39.2 | |
| Coding Index | 38.7 | |
| GPQA | 857% | |
| HLE | 227% | |
| SciCode | 43.4% | |
| IFBench | 75.6% | |
| LCR | 62% | |
| TerminalBench Hard | 36.4% | |
| Tau2 | 59.9% |
Data: Artificial Analysis · Updated: April 10, 2026
Frequently Asked Questions (15)
When was Gemma 4 31B (Reasoning) released?
Gemma 4 31B (Reasoning) was released on April 2, 2026.
Who created Gemma 4 31B (Reasoning)?
Gemma 4 31B (Reasoning) was created by Google.
How intelligent is Gemma 4 31B (Reasoning)?
Gemma 4 31B (Reasoning) scores 39 on the Artificial Analysis Intelligence Index, placing it well above average among other open weight models of similar size (median: 15).
How fast is Gemma 4 31B (Reasoning)?
Gemma 4 31B (Reasoning) generates output at 35.9 tokens per second (based on Google's API), which is at the lower end compared to other open weight models of similar size (median: 96.9 t/s).
What is the latency of Gemma 4 31B (Reasoning)?
Gemma 4 31B (Reasoning) has a time to first token (TTFT) of 1.70s (based on Google's API), which is better than average compared to other open weight models of similar size (median: 1.90s).
How verbose is Gemma 4 31B (Reasoning)?
When evaluated on the Intelligence Index, Gemma 4 31B (Reasoning) generated 39M output tokens, which is somewhat higher than average compared to other open weight models of similar size (median: 23M).
Is Gemma 4 31B (Reasoning) a reasoning model?
Yes, Gemma 4 31B (Reasoning) is a reasoning model. It uses extended thinking or chain-of-thought reasoning to work through complex problems before providing an answer.
What input modalities does Gemma 4 31B (Reasoning) support?
Gemma 4 31B (Reasoning) supports text, image, and video input.
What output modalities does Gemma 4 31B (Reasoning) support?
Gemma 4 31B (Reasoning) supports text output.
Can Gemma 4 31B (Reasoning) process images?
Yes, Gemma 4 31B (Reasoning) supports image input and can analyze, describe, and answer questions about images.
Is Gemma 4 31B (Reasoning) multimodal?
Yes, Gemma 4 31B (Reasoning) is multimodal. It can process text, image, and video input and generate text output.
What is the context window of Gemma 4 31B (Reasoning)?
Gemma 4 31B (Reasoning) has a context window of 260k tokens. This determines how much text and conversation history the model can process in a single request.
Is Gemma 4 31B (Reasoning) open source?
Yes, Gemma 4 31B (Reasoning) is open weights. The model weights are publicly available and can be downloaded for self-hosting.
How many parameters does Gemma 4 31B (Reasoning) have?
Gemma 4 31B (Reasoning) has 30.7 billion parameters.
What is the license for Gemma 4 31B (Reasoning)?
Gemma 4 31B (Reasoning) is released under the Apache 2.0 license. This license allows commercial use.
