Gemma 4 31B Instruct is a 30.7B dense model with a long context window, multilingual performance, function calling, and configurable reasoning. Running inside a TEE (Trusted Execution Environment), with provider attestation support.
Added May 26, 2026
Model weightsContext Window
262.1K
Max Output
262.1K
Avg output tokens (7d)
8.1K tokens
Input Price (Auto)
$0.15/1M
Output Price (Auto)
$0.46/1M
Cache Read (Auto)
$0.075/1M
Capabilities
Benchmarks
Benchmarks
Performance metrics and benchmarks
Sourced from Artificial Analysis.
Intelligence Index
24.8
Coding Index
43.4
Reasoning
GPQA Diamond
Graduate-level scientific reasoning
76.3%
Better than 68% of models compared
HLE
Humanity's Last Exam
11.5%
Better than 70% of models compared
IFBench
Instruction-following benchmark
53.5%
Better than 65% of models compared
T²-Bench Telecom
Conversational AI agents in dual-control scenarios
65.5%
Better than 60% of models compared
AA-LCR
Long context reasoning evaluation
36.0%
Better than 51% of models compared
Coding
SciCode
Python programming for scientific computing
41.1%
Better than 82% of models compared
Terminal-Bench Hard
Agentic coding and terminal use
30.3%
Better than 74% of models compared
Last updated Jun 28, 2026
Artificial AnalysisProviders
Auto routing is available for this model. Explicit provider selection is not available.
Loading provider options…
Related text models
Compare Gemma 4 31B IT TEE with similar models from the same provider or model family.
Gemma 4 31B Thinking TEE
TEE/gemma4-31b:thinkingGoogle's Gemma 4 31B instruction-tuned model in a Trusted Execution Environment (TEE) with thinking explicitly enabled and provider attestation support.
Gemma 4 31B
TEE/gemma4-31bGoogle's Gemma 4 31B instruction-tuned model in a Trusted Execution Environment (TEE) with attestation support. This route keeps reasoning disabled for faster direct answers while preserving structured output support.
Gemma 4 26B A4B Uncensored TEE
TEE/gemma-4-26b-a4b-uncensoredGemma 4 26B A4B Uncensored Heretic is a multimodal MoE model tuned to reduce refusal behavior while preserving coding, reasoning, and function-calling strengths. Running inside a TEE (Trusted Execution Environment), with provider attestation support.
Gemma 4 31B Agares v1
Gemma-4-31B-Agares-v1Gemma 4 31B Agares v1 is a community creative finetune for reasoning, multimodal chat, expressive writing, and roleplay.
Gemma 4 31B Animus V14.1
Gemma-4-31B-Animus-V14.1Gemma 4 31B Animus V14.1 is a community creative finetune for reasoning, multimodal chat, expressive writing, and roleplay.
Gemma 4 31B AssGuard
Gemma-4-31B-AssGuardGemma 4 31B AssGuard is a community creative finetune for reasoning, multimodal chat, expressive writing, and roleplay.