Skip to content

Google · Chat

Gemma 4 26B

Google 🇺🇸ChatGemini familyOpen weights. One of its sources says the weights are publicly available.
From Fastrouter's listing:
Gemma 4 26B is a 31B-parameter dense multimodal model from Google DeepMind, handling text, image, and audio input to generate text output. It supports an explicit thinking mode via control tokens, tool use, and a 256K-token context window, and is released under an Apache 2.0 license.
Max context
262K
Listings
5
Providers
4

Evidence and sources

Sources: Darkbloom data docs, FastRouter data, AKI data, Models.dev dataPrimary data from Darkbloom from FastRouter: name, description from AKI: family, variant from Models.dev: weights availability

Advantages and trade-offs

Compared with 10 representative chat models

Advantages

Open weights available

The weights are public. 4 of the 10 peers that say either way have open weights.

We only make a claim when there's enough data on models of the same type. Updated 22 Nov, 06:07 UTC.Prices come from the model's maker where they sell it, otherwise their official partner, otherwise the market median. This comparison uses 7 maker's prices.

Input & output

as stated by the model's providers
Text
Images
Text

Capabilities

Supported operations and provider-stated features

ReasoningFunction callingStructured output

Listings

 Googlegoogle-gemma-4-26bSource: Google Vertex AI PricingNot published by the provider$0.15cache read $0.015$0.60
 Fastroutergoogle/gemma4-26b:free · freeSource: FastRouter data262K$0.00$0.00
 Akigemma4-26bSources: AKI data, Models.dev data256K$0.10$0.50
 Darkbloomgemma-4-26bSource: Darkbloom data docsNot published by the provider$0.05cache read $0.0125$0.19
 Darkbloomgemma-4-26b-8bit · 8bitSource: Darkbloom data docsNot published by the provider$0.05cache read $0.0125$0.19

Benchmarks

Named effort levels run from lowest to highest. Select a result to see its source details.

Multilingual

IdrockBench Uzbek4 related benchmarks · 4 results

Instruction following

Reasoning

Data credits: IdrockBench, IDROCK AI Excellence Center, New Uzbekistan University. Public numeric results; no results licence stated (code MIT and dataset licences not borrowed).

Around the web

  • Gemma — Google DeepMind
    DolphinGemma DolphinGemma Democratizing AI with offline access Lentera runs an offline AI microserver for educators …DolphinGemma DolphinGemma Democratizing AI with offline access Lentera runs an offline AI microserver for educators and students Gemma 3n for Impact Developing sovereign AI Crane AI Labs improves Swa…

    ai.google.dev

  • Gemma models overview  |  Google AI for Developers
    Gemma models overview Alt Gemma logo Gemma is a family of lightweight, state-of-the-art open models …Gemma models overview Alt Gemma logo Gemma is a family of lightweight, state-of-the-art open models built from the same research and technology used to create the Gemini models. Developed by Google D…

    ai.google.dev

  • Use Gemma open models  |  Gemini Enterprise Agent Platform  |  Google Cloud Documentation
    Gemma is a set of lightweight, generative artificial intelligence (AI) open models. Gemma models are …Gemma is a set of lightweight, generative artificial intelligence (AI) open models. Gemma models are available to run in your applications and on your hardware, mobile devices, or hosted services. Yo…

    docs.cloud.google.com

  • Gemma Scope  |  Google AI for Developers
    Chatbot safety & debugging Analyze specific chat behaviors, refusal mechanisms, and chain-of-thought faithfulness to build …Chatbot safety & debugging Analyze specific chat behaviors, refusal mechanisms, and chain-of-thought faithfulness to build safer AI agents. Learn more Read the Gemma Scope 2 Blog Read about the new a…

    ai.google.dev

  • Gemma model fine-tuning  |  Google AI for Developers
    Gemma model fine-tuning Fine-tuning a generative artificial intelligence (AI) model such as Gemma modifies the …Gemma model fine-tuning Fine-tuning a generative artificial intelligence (AI) model such as Gemma modifies the behavior of the model. You typically fine-tune Gemma with the intent of improving its pe…

    ai.google.dev

  • Gemma explained: What’s new in Gemma 2 - Google Developers Blog
    NVIDIA TensorRT-LLM, as well as our JAX and Keras, enables optimization of performance and efficient …NVIDIA TensorRT-LLM, as well as our JAX and Keras, enables optimization of performance and efficient deployment across diverse hardware configurations. Here’s the core parameters of the new models: C…

    developers.googleblog.com

  • Deploy Gemma with Google Cloud  |  Google AI for Developers
    code samples are licensed under the Apache 2.0 License . For details, see the Google …code samples are licensed under the Apache 2.0 License . For details, see the Google Developers Site Policies . Java is a registered trademark of Oracle and/or its affiliates. Last updated 2026-07-02…

    ai.google.dev

  • GitHub - google-deepmind/gemma: Gemma open-weight LLM library, from Google DeepMind · GitHub
    cases, see our documentation . Please report issues and feedback in our GitHub . Installation …cases, see our documentation . Please report issues and feedback in our GitHub . Installation Install JAX for CPU, GPU or TPU. Follow the instructions on the JAX website . Run pip install gemma Examp…

    github.com

  • Use Gemma open models  |  Gemini Enterprise Agent Platform  |  Google Cloud Documentation
    requirements. To learn more, see the following tutorials in the GKE documentation: Serve Gemma with …requirements. To learn more, see the following tutorials in the GKE documentation: Serve Gemma with vLLM Serve Gemma with TGI Serve Gemma with Triton and TensorRT-LLM Serve Gemma with JetStream Model…

    docs.cloud.google.com

  • What Is Google Gemma? | IBM
    Gemma is Google’s family of free and open small language models (SLMs) . They’re built …Gemma is Google’s family of free and open small language models (SLMs) . They’re built from the same technology as the Gemini family of large language models (LLMs) and are considered “lightweight” v…

    ibm.com

Also known as: gemma-4-26b, gemma-4-26b-8bit, google-gemma-4-26b, google/gemma4-26b:free

Like this one. Tap or hover any dotted value, label or icon to see where it comes from and how it is measured. open details when tapped or hovered. How this data works