Skip to content

Gemma 4 26B A4B

google/gemma-4-26b-a4b-it

context
262K
modalities
image, text, video → text
capabilities
Tool use, Reasoning, Vision, Structured output, Low cost

Overview

Gemma 4 26B A4B IT is an instruction-tuned Mixture-of-Experts (MoE) model from Google DeepMind. Despite 25.2B total parameters, only 3.8B activate per token during inference — delivering near-31B quality at...

Providers

Access method depends on provider and account setup.

Pricing

  • Cloudflare

    input / 1M
    $0.10
    output / 1M
    $0.30
  • CoreWeave

    input / 1M
    $0.10
    output / 1M
    $0.30
  • Darkbloom

    input / 1M
    $0.042
    output / 1M
    $0.22
  • DeepInfra

    input / 1M
    $0.07
    output / 1M
    $0.34
  • DekaLLM

    input / 1M
    $0.06
    output / 1M
    $0.33
  • Google

    input / 1M
    $0.15
    output / 1M
    $0.60
  • Io Net

    input / 1M
    $0.11
    output / 1M
    $0.32
  • Makora

    input / 1M
    $0.10
    output / 1M
    $0.28
  • NextBit

    input / 1M
    $0.09
    output / 1M
    $0.30
  • Novita

    input / 1M
    $0.13
    output / 1M
    $0.40
  • Parasail

    input / 1M
    $0.13
    output / 1M
    $0.40
  • Reka

    input / 1M
    $0.06
    output / 1M
    $0.20
  • SiliconFlow

    input / 1M
    $0.14
    output / 1M
    $0.40
  • Venice

    input / 1M
    $0.13
    output / 1M
    $0.40

Use this model through Nautili

Route requests to Gemma 4 26B A4B through your Nautili gateway using a supported SDK or the Nautili SDK.

Connect your stack to Nautili

Route across supported providers through one control layer.