Google Gemma 4 31B: dense open-weights model built from Gemini 3 research, designed for consumer GPUs and workstations, with built-in thinking, MTP, and fine-tuning support.
Capabilities, design details, and architectural traits
Gemma 4 31B is Google's dense model in the Gemma 4 family, explicitly positioned to bridge server-grade performance and local execution on consumer GPUs and workstations. Within the Gemma 4 lineup, it is the variant that maximizes raw quality and serves as the primary foundation for fine-tuning, contrasting with the 26B MoE variant which optimizes for inference latency. The entire Gemma 4 family is built from Gemini 3 research and released as open weights.
| Trait | Detail |
|---|---|
| Architecture | Dense 31B - not Mixture-of-Experts; all parameters active during inference, prioritizing raw output quality over speed |
| Deployment target | Consumer GPUs and workstations; explicitly positioned for local-first AI development and fine-tuning |
| Multi-Token Prediction (MTP) | Supported; universally recommended for all tasks on GPU backends for this model size |
| Built-in thinking mode | Step-by-step reasoning before answering, available across the Gemma 4 family |
| Quantization checkpoints | QAT checkpoints available in compressed-tensors (-w4a16-ct) format for optimized cloud serving |
All Gemma 4 models, including the 31B, use a hybrid attention mechanism that interleaves local sliding window attention with full global attention. The final layer is always global. Global layers use unified Keys and Values with Proportional RoPE (p-RoPE) for long-context memory optimization.
Independent evaluations · Artificial Analysis
Evaluate specifications, pricing, and independent benchmark indices
| Model Details | |||
|---|---|---|---|
| General Info | |||
| Provider | Anthropic | Anthropic | |
| Release Date | April 2, 2026 | July 24, 2026 | June 9, 2026 |
| Knowledge Cutoff | - | May 2026 | - |
| Context & Limits | |||
| Context Window | 262K | 1M Best Context Window | 1M Best Context Window |
| Pricing (per 1M tokens) | |||
| Input Pricing | $0.15 Best Input Pricing | $5 | $10 |
| Output Pricing | $0.40 Best Output Pricing | $25 | $50 |
| Modalities | |||
| Inputs | imagetextvideo | textimage | textimagefile |
| Outputs | text | text | text |
| Benchmarks (0-100) | |||
| Intelligence Index | 22.3 | 63.1 Best Intelligence Index | 62.1 |
| Coding Index | 33.2 | 78.0 Best Coding Index | 76.5 |
| Agentic Index | 11.1 | 59.2 Best Agentic Index | 56.6 |
Graduate-level reasoning and expert Q&A evaluation.
Extremely difficult logical reasoning and knowledge.
Logical reasoning over long context windows.
Independent evaluation data provided by Artificial Analysis. To view the latest benchmarks and full details, visit their official site.
