Gemma 3 27B
Google · Multimodal · Released Mar 2025
Gemma 3 27B is Google's 27-billion-parameter open-weight text model with a 131K token context window.
- Strengths
- The 27B parameter count provides strong performance on complex reasoning and multi-step tasks while remaining practical for deployment, and the 131K context window handles long documents and extended conversations.
- Best for
- Applications requiring open-weight models with substantial reasoning capability, including local deployment, fine-tuning, and serverless inference at moderate scale.
- Limitations
- As an open-weight model it lacks the advanced reasoning capabilities of Google's Gemini reasoning models, and its performance is superseded by the newer Gemini 2.5 family for cloud-hosted inference workloads.
Input / 1M
$0.08
Output / 1M
$0.35
Cached input / 1M
$0.04
Context window
131K
Price history
Input (solid)Output (dashed)
Price change
- 30d
- in decreased 0.0% out decreased 22.2%
- 90d
- in decreased 0.0% out decreased 22.2%
- 1y
- in decreased 0.0% out increased 118.8%
- Since launch
- in decreased 0.0% out increased 118.8%
Snapshots
| Effective | Input | Output | Cached in | Note | Source |
|---|---|---|---|---|---|
| 18 Oct 2026 | $0.08 | $0.35 | $0.04 | Imported from OpenRouter | openrouter.ai |
| 23 Jul 2026 | $0.08 | $0.45 | $0.04 | Imported from OpenRouter | openrouter.ai |
| 17 Jul 2026 | $0.1 | $0.3 | — | Imported from OpenRouter | openrouter.ai |
| 15 Jul 2026 | $0.08 | $0.45 | $0.04 | Imported from OpenRouter | openrouter.ai |
| 11 Jun 2026 | $0.08 | $0.16 | — | Imported from OpenRouter | openrouter.ai |