NOUMENOpen app
NOUMEN

Google · Open weights

Gemma 4 31B Thinking TEE

Added May 2, 2026Open weights

Compare models

Choose up to 3 models to compare.1 of 3 chosen

Loading models…

Use in the app

Google's Gemma 4 31B instruction-tuned model in a Trusted Execution Environment (TEE) with thinking explicitly enabled and provider attestation support.

  • Reasoning
  • Tools
  • Vision
  • Structured output
Input
Text, Images
Output
Text
Discovery tier
Open weights

Pricing

USD per 1M tokens
Input
$0.40
Output
$1.00
Cache read
$0.40

Reference prices as published by the model provider. API example

Specifications

Context window
262.1K
Max output
131.1K
Average output (7 days)
1.1K tokens
Longer than 75% of 468 models

Side by side

Gemma 4 31B Thinking TEE compared

Specifications, prices, speed and benchmark scores next to each other. Benchmarks below show every compared model too.

Compare models

Choose up to 3 models to compare.1 of 3 chosen

Loading models…

Gemma 4 31B Thinking TEEGoogleThis modelDiffusionGemmaGoogleRemove
ProviderGoogleGoogle
Discovery tierOpen weightsOpen weights
AddedMay 2, 2026Sep 19, 2026
Limits
Context window262.14K262.14K
Max output131.07KBest32.77K
Average output1.1K tokens266 tokens
Pricing, USD per 1M tokens
Input$0.40$0.05Best
Output$1.00$0.15Best
Cache read$0.40$0.025Best
Performance
Request success (24h)80.3%—
Benchmark indexes
Intelligence Index14.7Best9.5
Coding Index43.4Best19.7
Capabilities
ReasoningYesYes
ToolsYesNo
VisionYesNo
Structured outputYesNo
Reasoning effort—none, xhigh

Public evaluations

Benchmarks

Independent evaluation results, with how each score ranks among the models in the Noumen catalog.

  • Gemma 4 31B Thinking TEE
  • DiffusionGemma

Intelligence Index

  • Gemma 4 31B Thinking TEE14.7
  • DiffusionGemma9.5

Gemma 4 31B Thinking TEE: Better than 35% of 326 scored models

3.8Catalog models by score55.1

Indexes

  • Coding Index

    Composite of the coding evaluations

    • Gemma 4 31B Thinking TEE: 43.4Top
    • DiffusionGemma: 19.7

Agentic work

  • AutomationBench-AA

    Workflow automation, penalized for guardrail violations

    • Gemma 4 31B Thinking TEE: 4.9%
    • DiffusionGemma: —

  • AutomationBench-AA tasks completed

    Workflows finished end to end without a guardrail violation

    • Gemma 4 31B Thinking TEE: 0.5%
    • DiffusionGemma: —

  • AA-Briefcasev1.1

    Agentic knowledge work, rated head to head

    • Gemma 4 31B Thinking TEE: 368 Elo
    • DiffusionGemma: —

  • GDPval-AAv2.1

    Economically valuable occupational tasks, rated head to head

    • Gemma 4 31B Thinking TEE: 619 EloTop
    • DiffusionGemma: 321 Elo

Document reasoning

  • GDP.pdf

    Professional reasoning over PDF documents, all-pass rate

    • Gemma 4 31B Thinking TEE: 6.0%
    • DiffusionGemma: —

  • AA-LCRv1.1

    Reasoning across long documents

    • Gemma 4 31B Thinking TEE: 69.7%Top
    • DiffusionGemma: 19.7%

Reasoning and knowledge

  • Humanity’s Last Exam

    Expert-level questions across academic fields

    • Gemma 4 31B Thinking TEE: 23.6%Top
    • DiffusionGemma: 10.8%

  • GPQA Diamond

    Graduate-level science questions

    • Gemma 4 31B Thinking TEE: 85.7%Top
    • DiffusionGemma: 66.9%

  • IFBench

    Following precise output instructions

    • Gemma 4 31B Thinking TEE: 75.6%Top
    • DiffusionGemma: 59.5%

Coding

  • Terminal-Benchv4.0

    Practical software engineering in a terminal

    • Gemma 4 31B Thinking TEE: 0.0%
    • DiffusionGemma: —

  • SciCode

    Scientific computing in Python

    • Gemma 4 31B Thinking TEE: 45.5%
    • DiffusionGemma: —

Legacy evaluations

  • Terminal-Bench Hard

    Earlier terminal task suite

    • Gemma 4 31B Thinking TEE: 36.4%
    • DiffusionGemma: —

  • τ²-Bench Telecom

    Tool use in a customer-support simulation

    • Gemma 4 31B Thinking TEE: 59.9%
    • DiffusionGemma: —

  • AA-LCR (unversioned)

    Long-context reasoning, earlier grading

    • Gemma 4 31B Thinking TEE: 69.7%Top
    • DiffusionGemma: 19.7%

Scores from Artificial Analysis, updated Oct 26, 2026. Artificial Analysis

Live routing data

Performance and reliability

Recent averages across requests routed for this model. Your own speed depends on prompt length, reasoning effort and load.

Prompt cache hits
30.1%
About $0.40 per 1M input tokens with caching
Request success (24h)
80.3%
Unstable

Cache hits are the share of input tokens recently served from a prompt cache across all requests. They are not a forecast of your own hit rate.

Hourly request success · last 72 hoursLowest hour 33.3%
  • Healthy (98% or more)
  • Degraded (90–98%)
  • Down (under 90%)
  • No requests

Developers

Use it through the API

Noumen’s API is OpenAI-compatible: any Chat Completions client works with a base URL and a key.

Read the quickstart
curl https://www.noumen.xyz/api/v1/chat/completions \
  -H "Authorization: Bearer $NOUMEN_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "TEE/gemma4-31b:thinking",
    "messages": [{ "role": "user", "content": "Explain this stack trace." }]
  }'
import os
from openai import OpenAI

client = OpenAI(base_url="https://www.noumen.xyz/api/v1", api_key=os.environ["NOUMEN_API_KEY"])
reply = client.chat.completions.create(
    model="TEE/gemma4-31b:thinking",
    messages=[{"role": "user", "content": "Explain this stack trace."}],
)
print(reply.choices[0].message.content)

Google

Other Google models, starting with the same family.