Gemma 4 26B A4B

Google · google/gemma-4-26b-a4b-it

← Back to leaderboard

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...

Gemma 4 26B A4B ranks #46 of 348 models in the default AI Benchmark Hub order (quality first, then cost). At $0.20 blended per million tokens, it is cheaper than 81% of models with published API pricing. Its 262,144-token context window is about the catalog median (262,144 tokens). A quality index of 72.5 places it at #46 of 59 models with an Arena Elo signal.

open weightsimagetexttext+image+video->textvideo

Context

Max context: 262144
Max output: 16384

Pricing

Input / 1M: 0.07
Output / 1M: 0.34
Blend / 1M: 0.20

Quality

Quality index: 72.5

Provider

Provider: Google
Moderated: no

Monthly cost at this blended rate

Estimated API spend if every token is billed at $0.20per million (the blend of this model's input and output prices). Real bills depend on the input/output mix.

Tokens / monthEst. cost (USD)
1,000,000$0.20
10,000,000$2.05
100,000,000$20.50

Related models

Same lab, similar price, or similar quality — useful next comparisons.

More from Google

Closest API price

Closest quality index