Google: Gemma 3 12B
Analysis Summary
Google's Gemma 3 12B is a mid-small open-weight model released in March 2025, supporting text and image input with tool use and function calling. Its intelligence index of 8.8 and coding index of 6.3 place it in the limited-to-moderate tier, though its math index of 18.3 and MMLU-Pro of 0.595 show reasonable general knowledge breadth. The 131K context window is adequate for most document tasks.
For businesses, the combination of vision, tool use, and function calling at $0.05 input and $0.15 output per million tokens makes it an attractive option for cost-sensitive automation pipelines. It can handle structured extraction, basic tool-calling workflows, and image-aware tasks without significant cost overhead. Instruction-following scores are moderate, so prompt engineering matters.
It is best positioned as a budget-friendly component in multi-model architectures, handling routine structured tasks while more capable models handle complex reasoning. Teams building high-volume, low-complexity pipelines with tool integration will find it a practical choice.
Assessed June 9, 2026
Editorial notes
Gemma 3 12B from Google offers vision, tool use, and function calling at very low cost, with moderate reasoning capability suitable for lightweight business automation.
Rankings consider pricing, capabilities, benchmarks, and real-world applicability and are refreshed as new models launch. Feedback?
Performance Profile
How Google: Gemma 3 12B compares
Google: Gemma 3 12B ranks #333 of 377 AI models we track for overall intelligence, #269 of 314 for coding, #279 of 289 for agentic tasks. Its 131K-token context window is larger than 59% of the models we list. At $0.05 per million input tokens it is cheaper than 73% of comparable models.
About Google: Gemma 3 12B
Gemma 3 introduces multimodality, supporting vision-language input and text outputs. It handles context windows up to 128k tokens, understands over 140 languages, and offers improved math, reasoning, and chat capabilities,..
Capabilities
Architecture Detail
| Instruct Type | gemma |
Performance Indices
Source: Artificial Analysis
Benchmark Scores
Intelligence
Technical
Content
Benchmark data from Artificial Analysis and Hugging Face
How does Google: Gemma 3 12B stack up?
Compare side-by-side with other professional models.
Model Information
| OpenRouter ID |
google/gemma-3-12b-it
|
| Provider | |
| Release Date | March 13, 2025 |
| Context Length | 131,072 tokens |
| Max Completion | 16,384 tokens |
| Status | Active |
Pricing
| Token Type | Cost per 1M tokens | Cost per 1K tokens |
|---|---|---|
| Input | $0.05 | $0.000050 |
| Output | $0.15 | $0.000150 |
Live Performance
Live endpoint metrics, refreshed every 30 minutes.
Leaderboard Categories
External Resources
Frequently asked questions about Google: Gemma 3 12B
How much does Google: Gemma 3 12B cost?
Google: Gemma 3 12B costs $0.05 per million input tokens and $0.15 per million output tokens.
What is the context window of Google: Gemma 3 12B?
Google: Gemma 3 12B has a context window of 131,072 tokens (131K).
Is Google: Gemma 3 12B good for coding?
On our coding benchmark index, Google: Gemma 3 12B ranks #269 of 314 models, placing it in the broader range of the field for code generation and debugging.
What can Google: Gemma 3 12B do?
Google: Gemma 3 12B supports image/vision input, tool use, and function calling.
Who created Google: Gemma 3 12B?
Google: Gemma 3 12B is developed by Google and was released on March 13, 2025.
