NOUMENOpen app
NOUMEN

Google · Open weights

Gemma 4 31B IT TEE

Added May 26, 2026Open weights

Compare models

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

Loading models…

Use in the app

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.

  • Reasoning
  • Tools
Input
Text
Output
Text
Discovery tier
Open weights

Pricing

USD per 1M tokens
Input
$0.15
Output
$0.46

Reference prices as published by the model provider. API example

Specifications

Context window
262.1K
Max output
262.1K
Average output (7 days)
269 tokens
Longer than 28% of 465 models

Side by side

Gemma 4 31B IT 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 IT TEEGoogleThis modelQwen3.6 35B A3B TEEQwenRemove
ProviderGoogleQwen
Discovery tierOpen weightsOpen weights
AddedMay 26, 2026Jun 9, 2026
Limits
Context window262.14K262.14K
Max output262.14K262.14K
Average output269 tokens3.9K tokens
Pricing, USD per 1M tokens
Input$0.15Best$0.20
Output$0.46Best$1.27
Performance
Throughput53.0 tok/s—
Time to first token5.4 s—
Request success (24h)99.8%Best99.5%
Benchmark indexes
Intelligence Index14.715.2Best
Coding Index43.4Best28.1
Capabilities
ReasoningYesNo
ToolsYesNo
Reasoning effort——

Public evaluations

Benchmarks

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

  • Gemma 4 31B IT TEE
  • Qwen3.6 35B A3B TEE

Intelligence Index

  • Gemma 4 31B IT TEE14.7
  • Qwen3.6 35B A3B TEE15.2

Gemma 4 31B IT 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 IT TEE: 43.4Top
    • Qwen3.6 35B A3B TEE: 28.1

Agentic work

  • AutomationBench-AA

    Workflow automation, penalized for guardrail violations

    • Gemma 4 31B IT TEE: 4.9%
    • Qwen3.6 35B A3B TEE: —

  • AutomationBench-AA tasks completed

    Workflows finished end to end without a guardrail violation

    • Gemma 4 31B IT TEE: 0.5%
    • Qwen3.6 35B A3B TEE: —

  • AA-Briefcasev1.1

    Agentic knowledge work, rated head to head

    • Gemma 4 31B IT TEE: 368 Elo
    • Qwen3.6 35B A3B TEE: —

  • GDPval-AAv2.1

    Economically valuable occupational tasks, rated head to head

    • Gemma 4 31B IT TEE: 619 Elo
    • Qwen3.6 35B A3B TEE: 847 EloTop

Document reasoning

  • GDP.pdf

    Professional reasoning over PDF documents, all-pass rate

    • Gemma 4 31B IT TEE: 6.0%
    • Qwen3.6 35B A3B TEE: —

  • AA-LCRv1.1

    Reasoning across long documents

    • Gemma 4 31B IT TEE: 69.7%Top
    • Qwen3.6 35B A3B TEE: 64.3%

Reasoning and knowledge

  • Humanity’s Last Exam

    Expert-level questions across academic fields

    • Gemma 4 31B IT TEE: 23.6%Top
    • Qwen3.6 35B A3B TEE: 13.9%

  • GPQA Diamond

    Graduate-level science questions

    • Gemma 4 31B IT TEE: 85.7%Top
    • Qwen3.6 35B A3B TEE: 81.7%

  • IFBench

    Following precise output instructions

    • Gemma 4 31B IT TEE: 75.6%Top
    • Qwen3.6 35B A3B TEE: 36.2%

Coding

  • Terminal-Benchv4.0

    Practical software engineering in a terminal

    • Gemma 4 31B IT TEE: 0.0%
    • Qwen3.6 35B A3B TEE: —

  • SciCode

    Scientific computing in Python

    • Gemma 4 31B IT TEE: 45.5%
    • Qwen3.6 35B A3B TEE: —

Legacy evaluations

  • Terminal-Bench Hard

    Earlier terminal task suite

    • Gemma 4 31B IT TEE: 36.4%Top
    • Qwen3.6 35B A3B TEE: 25.8%

  • τ²-Bench Telecom

    Tool use in a customer-support simulation

    • Gemma 4 31B IT TEE: 59.9%
    • Qwen3.6 35B A3B TEE: 85.1%Top

  • AA-LCR (unversioned)

    Long-context reasoning, earlier grading

    • Gemma 4 31B IT TEE: 69.7%Top
    • Qwen3.6 35B A3B TEE: 64.3%

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.

Throughput
53.0 tok/s
Faster than 38% of 250 models
Time to first token
5.4 s
Quicker than 30% of 250 models
Prompt cache hits
29.6%
Request success (24h)
99.8%
Reliable

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 96.1%
  • 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/gemma-4-31b-it",
    "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/gemma-4-31b-it",
    messages=[{"role": "user", "content": "Explain this stack trace."}],
)
print(reply.choices[0].message.content)

Google

Other Google models, starting with the same family.