Gemma 4 12B
Encoder-free multimodal mid-size open model with a 256K window — the sweet spot for single-GPU self-hosting; download-only (not exposed on the hosted Gemini API).
Data checked 2026-09-06Google DeepMind ga open weights
- Context window
- 262.1K tokens in
- Max output
- —
- Input
- $0.00 per 1M tokens
- Output
- $0.00 per 1M tokens
- Blended
- $0.00 3:1 in:out
- Released
- 2026
- Parameters
- 12B
- Licence
- Gemma 4 license (Gemma Terms of Use)
A What it is for
At a three-to-one input-to-output ratio, Gemma 4 12B costs $0.00 per million tokens blended. A workload of one million input and 330,000 output tokens per day would run about $0.00 per month at list price, before caching or batch discounts.
B Capabilities
| Input modalities | text, image |
|---|---|
| Output modalities | text |
C Others from Google DeepMind
| Model | Context | Max out | In $/M | Out $/M | Blended | Status |
|---|---|---|---|---|---|---|
Gemini 3.8 Flash
gemini-3.8-flash
|
1M | 65.5K | $0.75 | $3.75 | $1.50 | ga |
Gemini 3.7 Flash
gemini-3.7-flash
|
1M | 65.5K | $0.75 | $3.75 | $1.50 | ga |
Gemini 3.6 Flash
gemini-3.6-flash
|
1M | 65.5K | $0.75 | $3.75 | $1.50 | ga |
Gemini 3.5 Flash
gemini-3.5-flash
|
1M | 65.5K | $1.50 | $9.00 | $3.38 | ga |
Gemini 3.5 Flash-Lite
gemini-3.5-flash-lite
|
1M | 65.5K | $0.30 | $2.50 | $0.85 | ga |
Gemini 3.1 Flash-Lite
gemini-3.1-flash-lite
|
1M | 65.5K | $0.25 | $1.50 | $0.56 | ga |
Gemini 3.1 Pro Preview
gemini-3.1-pro-preview
|
1M | 65.5K | $2.00 | $12.00 | $4.50 | preview |
Gemini 3 Flash Preview
gemini-3-flash-preview
|
1M | 65.5K | $0.50 | $3.00 | $1.13 | preview |
Gemini 2.5 Pro
gemini-2.5-pro
|
1M | 65.5K | $1.25 | $10.00 | $3.44 | ga |
Gemini 2.5 Flash
gemini-2.5-flash
|
1M | 65.5K | $0.30 | $2.50 | $0.85 | ga |
Gemini 2.5 Flash-Lite
gemini-2.5-flash-lite
|
1M | 65.5K | $0.10 | $0.40 | $0.18 | ga |
Gemini 2.5 Computer Use Preview
gemini-2.5-computer-use-preview-10-2025
|
128K | 64K | $1.25 | $10.00 | $3.44 | preview |
D Verify before you commit
Model pricing changes without notice and this page is a snapshot. Confirm against the vendor's own page before you build a budget on it.
Source: ai.google.dev · checked 2026-09-06 · Google DeepMind pricing · API docs
Some figures on this page are recorded at medium confidence — the vendor does not publish them in a single authoritative place. Treat them as indicative.