Gemma 4 26B A4B
Google · Multimodal · Released Apr 2026
Google DeepMind's 26-billion-parameter open-weight model supporting text and image input with a 262K token context window.
- Strengths
- Offers a middle ground between inference speed and model capability as a smaller sibling to Gemma 4 31B, with efficient handling of longer documents within the 262K token window.
- Best for
- Applications requiring multimodal understanding (text and images) where parameter efficiency and reduced inference latency matter more than maximum reasoning depth.
- Limitations
- As an open-weight model, it typically trails Google's proprietary Gemini series in reasoning and instruction-following performance; the 26B parameter size places it below Gemma 4 31B for complex tasks.
Input / 1M
$0.042
Output / 1M
$0.22
Cached input / 1M
—
Context window
131K
Price history
Input (solid)Output (dashed)
Price change
- 30d
- in decreased 65.0% out decreased 45.0%
- 90d
- in decreased 30.0% out decreased 33.3%
- 1y
- in decreased 30.0% out decreased 33.3%
- Since launch
- in decreased 30.0% out decreased 33.3%
Snapshots
| Effective | Input | Output | Cached in | Note | Source |
|---|---|---|---|---|---|
| 10 Sep 2026 | $0.042 | $0.22 | — | Imported from OpenRouter | openrouter.ai |
| 17 Aug 2026 | $0.07 | $0.34 | — | Imported from OpenRouter | openrouter.ai |
| 10 Aug 2026 | $0.12 | $0.4 | $0.05 | Imported from OpenRouter | openrouter.ai |
| 29 Jul 2026 | $0.07 | $0.34 | — | Imported from OpenRouter | openrouter.ai |
| 28 Jul 2026 | $0.15 | $0.45 | $0.05 | Imported from OpenRouter | openrouter.ai |
| 27 Jul 2026 | $0.14 | $0.42 | $0.05 | Imported from OpenRouter | openrouter.ai |
| 23 Jul 2026 | $0.12 | $0.35 | $0.05 | Imported from OpenRouter | openrouter.ai |
| 19 Jul 2026 | $0.07 | $0.34 | — | Imported from OpenRouter | openrouter.ai |
| 15 Jul 2026 | $0.1 | $0.3 | — | Imported from OpenRouter | openrouter.ai |
| 11 Jun 2026 | $0.06 | $0.33 | — | Imported from OpenRouter | openrouter.ai |