modelgrep
G

Google: Gemma 4 26B A4B

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

75th smartest of 178Cheaper than 92% of paidReasoningToolsJSONVision
Use via OpenRouter ↗
Intelligence
31.2
75th of 178
Design Elo
Speed
68
115th fastest
Latency
356ms
first token
Input price
$0.060
23rd cheapest
Context
262K

How it compares

Smarter than58%
of all ranked models
Faster than61%
of all ranked models
Cheaper than92%
of all ranked models

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...

Benchmarks

independent · via OpenRouter
Artificial Analysis56th percentile
Intelligence Index
31.2
Coding Index
22.4
Agentic Index
32.1
GPQA Diamond
79%
Humanity's Last Exam
18%
SciCode
40%
Tau²-Bench (agentic)
44%

Providers & pricing (9)

ProviderIn $/MOut $/MUptime
DekaLLMbf16$0.060$0.33098%
DeepInfrafp8$0.070$0.34098.7%
Cloudflare$0.100$0.30099.2%
SiliconFlowfp8$0.120$0.40095.7%
Parasailbf16$0.130$0.40098.6%
Novitabf16$0.130$0.40050.4%
NextBitbf16$0.130$0.40098.4%
Google$0.150$0.60097.4%
Venicebf16$0.163$0.50097.5%

Specifications

Context window262K
Max output
Knowledge cutoff
Input modalitiesimage, text, video
Output modalitiestext
Prompt caching
Cache read price
ModeratedNo

Gemma 4 26B A4B FAQ

How much does Gemma 4 26B A4B cost?

Gemma 4 26B A4B costs $0.060 per million input tokens and $0.330 per million output tokens via OpenRouter, making it 23rd cheapest of 298 paid models.

How smart is Gemma 4 26B A4B ?

Gemma 4 26B A4B scores 31.2 on the Artificial Analysis Intelligence Index, ranking 75th of 178 benchmarked models, with a GPQA Diamond score of 79%.

How fast is Gemma 4 26B A4B ?

Gemma 4 26B A4B generates around 68 tokens per second with 356ms time-to-first-token (p50), the 115th fastest tracked model.

What is Gemma 4 26B A4B 's context window?

Gemma 4 26B A4B supports a 262K-token context window. It accepts image, text, video input.

Compare head-to-head