AI TOOLS RADAR
Submit
All models / GOOGLE
google logo
google

Google: Gemma 4 26B A4B

Source description (English)

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...

View source on OpenRouter Hugging Face
Overview

Model specifications

Context
262,144 tokens
Maximum output
235,929 tokens
Architecture
text+image+video->text
Tokenizer
Gemma
Knowledge cutoff
Not provided
Moderated
No
OPENROUTER

Complete pricing

Synchronized OpenRouter rates. Token prices are shown per one million tokens.

Input
$0.09
per 1M tokens
Output
$0.3
per 1M tokens
Cache read
$0.05
per 1M tokens
API

Model configuration

Default parameters

temperature
1
top_p
0.95
top_k
64

Reasoning

Reasoning required
No
Default parameters
No
API

Quick start

Set OPENROUTER_API_KEY locally. Python requires requests; JavaScript runs in Node.js. Keep the key on the server.

API documentation

curl --fail-with-body https://openrouter.ai/api/v1/chat/completions \
  -H "Authorization: Bearer $OPENROUTER_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"google/gemma-4-26b-a4b-it","messages":[{"role":"user","content":"Hello!"}]}'
Capabilities

Input → Output

Input
imagetextvideo
Output
text
Reasoning

No

API

Supported API parameters

frequency_penaltyinclude_reasoninglogit_biaslogprobsmax_tokensmin_ppresence_penaltyreasoningrepetition_penaltyresponse_formatseedstopstructured_outputstemperaturetool_choicetoolstop_ktop_logprobstop_p
Available providers

11 providers

Checked: September 13, 2026

Live providers on OpenRouter

Provider availability, latency, throughput and routing can change continuously. Open the source page for current operational data.

OpenRouter

Darkbloom

Not provided
Context
131K
Maximum output
33K
Input
$0.042
Output
$0.22
Cache read
Not provided
Cache write
Not provided

DekaLLM

bf16
Context
262K
Maximum output
236K
Input
$0.06
Output
$0.33
Cache read
Not provided
Cache write
Not provided

DeepInfra

fp8
Context
262K
Maximum output
16K
Input
$0.07
Output
$0.34
Cache read
Not provided
Cache write
Not provided

NextBit

bf16
Context
262K
Maximum output
236K
Input
$0.09
Output
$0.3
Cache read
$0.05
Cache write
Not provided

Cloudflare

Not provided
Context
256K
Maximum output
230K
Input
$0.1
Output
$0.3
Cache read
Not provided
Cache write
Not provided

Makora

Not provided
Context
262K
Maximum output
236K
Input
$0.1
Output
$0.34
Cache read
$0.034
Cache write
Not provided

Venice

bf16
Context
256K
Maximum output
8K
Input
$0.13
Output
$0.4
Cache read
$0.05
Cache write
Not provided

Parasail

bf16
Context
262K
Maximum output
236K
Input
$0.13
Output
$0.4
Cache read
$0.05
Cache write
Not provided

Novita

bf16
Context
262K
Maximum output
131K
Input
$0.13
Output
$0.4
Cache read
Not provided
Cache write
Not provided

SiliconFlow

fp8
Context
262K
Maximum output
236K
Input
$0.14
Output
$0.4
Cache read
$0.05
Cache write
Not provided

Google

Not provided
Context
262K
Maximum output
236K
Input
$0.15
Output
$0.6
Cache read
Not provided
Cache write
Not provided
google

4 models

All models
google logogoogle

Google: Gemini 3.8 Flash

Gemini 3.8 Flash is Google's most intelligent Flash model with significant gains from 3.7 Flash across software engineering, agentic tasks, and multi-step reasoning.

Context
1.0M
Input
text · image · video · file · audio
Output
text
Input: $0.75Output: $3.75per 1M tokens
View model details →
google logogoogle

Google: Gemini 3.8 Flash (batch)

Gemini 3.8 Flash is Google's most intelligent Flash model with significant gains from 3.7 Flash across software engineering, agentic tasks, and multi-step reasoning.

Context
1.0M
Input
text · image · video · file · audio
Output
text
Input: $0.375Output: $1.875per 1M tokens
View model details →
google logogoogle

Google: Gemini 3.7 Flash

Gemini 3.7 Flash is a multimodal model from Google for fast agentic workflows, coding, and complex multi-step reasoning. It is designed for tasks that require responsive performance and reliable multi-step...

Context
1.0M
Input
text · image · video · file · audio
Output
text
Input: $0.75Output: $3.75per 1M tokens
View model details →
google logogoogle

Google: Gemini 3.7 Flash (batch)

Gemini 3.7 Flash is a multimodal model from Google for fast agentic workflows, coding, and complex multi-step reasoning. It is designed for tasks that require responsive performance and reliable multi-step...

Context
1.0M
Input
text · image · video · file · audio
Output
text
Input: $0.375Output: $1.875per 1M tokens
View model details →