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...
Google: Gemma 4 26B A4B usage over time
OpenRouter share of tokens, by day.
Google: Gemma 4 26B A4B holds 0.30% today, up from 0.30% at the start of the visible window.
Rank history
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?
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.
Token rate sets the price; session length and reasoning set the bill.
| Agent | 1 turn | 2–9 turns | 10–49 turns |
|---|---|---|---|
| Hermes Agent | $0.0016 | $0.0050 | $0.037 |
Where to run it
Google: Gemma 4 26B A4B pricing by provider
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.
| Provider | Input $/M | Output $/M | Context | Uptime |
|---|---|---|---|---|
| Darkbloom | $0.04 | $0.22 | 131k | 99.83% |
| Cloudflare | $0.10 | $0.30 | 256k | 99.87% |
| DeepInfra | $0.07 | $0.34 | 262k | 98.63% |
| NextBit | $0.10 | $0.40 | 262k | 99.88% |
| SiliconFlow | $0.12 | $0.40 | 262k | 99.72% |
| Venice | $0.13 | $0.40 | 256k | 98.72% |
| Parasail | $0.13 | $0.40 | 262k | 98.90% |
| Novita | $0.13 | $0.40 | 262k | 99.22% |
| $0.15 | $0.60 | 262k | 99.44% |