Google

Google: Gemma 4 26B A4B

google/gemma-4-26b-a4b-it-20260403Hugging Face ↗

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...

toolsstructuredreasoningimage in
Data source

Google: Gemma 4 26B A4B usage over time

OpenRouter share of tokens, by day.

View
Tidelines — LLM usage and market share

Google: Gemma 4 26B A4B holds 0.30% today, up from 0.30% at the start of the visible window.

Data: OpenRouter as of 2026-09-01 · Source: tidelines.ai/models/google-gemma-4-26b-a4b-it-20260403

Rank history

Tidelines — LLM usage and market share
Data: OpenRouter as of 2026-09-01 · Source: tidelines.ai/models/google-gemma-4-26b-a4b-it-20260403

Adoption curve vs the pantheon

Google: Gemma 4 26B A4B 's share plotted against days since first tracked, overlaid on other current leaders on the same axis. The question every new-model page asks: is this one going to make it?

Tidelines — LLM usage and market share
Data: OpenRouter as of 2026-09-01 · Source: tidelines.ai/models/google-gemma-4-26b-a4b-it-20260403

Citable stat

As of 2026-09-01, Google: Gemma 4 26B A4B holds 0.3% of OpenRouter token share, down 0.34pp over 30 days. Source: Tidelines (OpenRouter).

Cost per session

Median dollars spent on one Google: Gemma 4 26B A4B session, by coding agent and session length.

Tidelines — LLM usage and market share

Token rate sets the price; session length and reasoning set the bill.

Agent1 turn2–9 turns10–49 turns
Hermes Agent$0.0016$0.0050$0.037

Compare every model's session cost →

Data: OpenRouter session-cost dataset as of 2026-08-30 · Source: tidelines.ai/models/google-gemma-4-26b-a4b-it-20260403

Where to run it

Google: Gemma 4 26B A4B pricing by provider

Tidelines — LLM usage and market share

Per-provider pricing, context, and uptime via OpenRouter. Sorted by output price.

Google: Gemma 4 26B A4B is served by 9 tracked endpoints on OpenRouter. The cheapest output price is $0.22 per million tokens from Darkbloom, and the most expensive is $0.60 from Google. The largest context window offered is 262k tokens. Prices are list rates and refresh daily.

ProviderInput $/MOutput $/MContextUptime
Darkbloom$0.04$0.22131k99.83%
Cloudflare$0.10$0.30256k99.87%
DeepInfra$0.07$0.34262k98.63%
NextBit$0.10$0.40262k99.88%
SiliconFlow$0.12$0.40262k99.72%
Venice$0.13$0.40256k98.72%
Parasail$0.13$0.40262k98.90%
Novita$0.13$0.40262k99.22%
Google$0.15$0.60262k99.44%