NOUMENOpen app
NOUMEN

Google · Open weights

DiffusionGemma

Added Sep 19, 2026Open weights

Compare models

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

Loading models…

Use in the app

DiffusionGemma is a high-speed diffusion-based version of Gemma 4 26B A4B. It supports optional reasoning and a 262,144-token context window. Temperature, min-p, seed, minimum output tokens, logit bias, bad words, and allowed token IDs are unavailable; omit these parameters to use native diffusion sampling.

  • Reasoning
Input
Text
Output
Text
Reasoning effort
none · xhigh
Discovery tier
Open weights

Pricing

USD per 1M tokens
Input
$0.05
Output
$0.15
Cache read
$0.025

Reference prices as published by the model provider. API example

Specifications

Context window
262.1K
Max output
32.8K
Average output (7 days)
266 tokens
Longer than 28% of 473 models

Side by side

DiffusionGemma 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…

DiffusionGemmaGoogleThis modelLlama 4 MaverickMetaRemove
ProviderGoogleMeta
Discovery tierOpen weightsOpen weights
AddedSep 19, 2026Sep 5, 2025
Limits
Context window262.14K1.05MBest
Max output32.77K65.54KBest
Average output266 tokens114 tokens
Pricing, USD per 1M tokens
Input$0.05Best$0.19
Output$0.15Best$0.65
Cache read$0.025Best$0.05
Performance
Request success (24h)—99.6%
Benchmark indexes
Intelligence Index9.5Best9.3
Coding Index19.7Best16.3
Math Index—19.3
Capabilities
ReasoningYesNo
ToolsNoYes
VisionNoYes
Structured outputNoYes
Reasoning effortnone, xhigh—

Public evaluations

Benchmarks

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

  • DiffusionGemma
  • Llama 4 Maverick

Intelligence Index

  • DiffusionGemma9.5
  • Llama 4 Maverick9.3

DiffusionGemma: Better than 19% of 326 scored models

3.8Catalog models by score55.1

Indexes

  • Coding Index

    Composite of the coding evaluations

    • DiffusionGemma: 19.7Top
    • Llama 4 Maverick: 16.3

  • Math Index

    Composite of the math evaluations

    • DiffusionGemma: —
    • Llama 4 Maverick: 19.3

Agentic work

  • AutomationBench-AA

    Workflow automation, penalized for guardrail violations

    • DiffusionGemma: —
    • Llama 4 Maverick: 0.2%

  • AutomationBench-AA tasks completed

    Workflows finished end to end without a guardrail violation

    • DiffusionGemma: —
    • Llama 4 Maverick: 0.0%

  • GDPval-AAv2.1

    Economically valuable occupational tasks, rated head to head

    • DiffusionGemma: 321 Elo
    • Llama 4 Maverick: —

Document reasoning

  • GDP.pdf

    Professional reasoning over PDF documents, all-pass rate

    • DiffusionGemma: —
    • Llama 4 Maverick: 0.8%

  • AA-LCRv1.1

    Reasoning across long documents

    • DiffusionGemma: 19.7%
    • Llama 4 Maverick: 50.0%Top

Reasoning and knowledge

  • Humanity’s Last Exam

    Expert-level questions across academic fields

    • DiffusionGemma: 10.8%Top
    • Llama 4 Maverick: 4.9%

  • GPQA Diamond

    Graduate-level science questions

    • DiffusionGemma: 66.9%
    • Llama 4 Maverick: 67.1%Top

  • MMLU-Pro

    Multi-subject knowledge and reasoning

    • DiffusionGemma: —
    • Llama 4 Maverick: 80.9%

  • IFBench

    Following precise output instructions

    • DiffusionGemma: 59.5%Top
    • Llama 4 Maverick: 43.0%

Coding

  • Terminal-Benchv4.0

    Practical software engineering in a terminal

    • DiffusionGemma: —
    • Llama 4 Maverick: 0.0%

  • SciCode

    Scientific computing in Python

    • DiffusionGemma: —
    • Llama 4 Maverick: 31.7%

  • LiveCodeBench

    Recent competitive programming problems

    • DiffusionGemma: —
    • Llama 4 Maverick: 39.7%

Math

  • AIME 2025

    Competition mathematics, 2025 problems

    • DiffusionGemma: —
    • Llama 4 Maverick: 19.3%

  • AIME

    Competition mathematics

    • DiffusionGemma: —
    • Llama 4 Maverick: 39.0%

  • MATH-500

    Competition math problems

    • DiffusionGemma: —
    • Llama 4 Maverick: 88.9%

Legacy evaluations

  • Terminal-Bench Hard

    Earlier terminal task suite

    • DiffusionGemma: —
    • Llama 4 Maverick: 6.8%

  • τ²-Bench Telecom

    Tool use in a customer-support simulation

    • DiffusionGemma: —
    • Llama 4 Maverick: 17.8%

  • AA-LCR (unversioned)

    Long-context reasoning, earlier grading

    • DiffusionGemma: 19.7%
    • Llama 4 Maverick: 50.0%Top

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
0.0%
About $0.05 per 1M input tokens with caching

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.

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": "google/diffusiongemma",
    "reasoning_effort": "none",
    "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="google/diffusiongemma",
    messages=[{"role": "user", "content": "Explain this stack trace."}],
)
print(reply.choices[0].message.content)

Google

Other Google models, starting with the same family.