Gemma 3 4B
by Merge Gateway · family gemma · listed Mar 2025VW
Open multimodal Gemma instruction model for efficient text generation and image understanding
Current rates USD per 1M tokens
Input$0.04
Output$0.08
Cache read—
Cache write—
Against the market
On the record
- Model id
google/gemma-3-4b-it- Context window
- 128K tokens
- Max output
- 8K tokens
- Input modalities
- text, image
- Output modalities
- text
- Reasoning
- no
- Tool calls
- no
- Open weights
- yes
- Knowledge cutoff
- Aug 2024
- Released
- 2025-03-12
- Last updated
- 2025-03-12
- Provider docs
- docs.merge.dev/merge-gateway
Back of the envelope
≈
Also listed at
- Nvidiafree in · free out
- Deep Infra$0.05 in · $0.10 out
- Kilo Gateway$0.05 in · $0.10 out
- OpenRouter$0.05 in · $0.10 out
- Hugging Face$0.05 in · $0.10 out
- Eden AI$0.05 in · $0.10 out
- NanoGPT$0.201 in · $0.201 out