Google › Gemma 3 4B › Pricing
$0.04 / $0.05per 1M tokens in / out
$0.02cache read · −50%
#321on OpenRouter by tokens · 1.2B / week
Price sheetper 1M tokens
| Input | $0.04 | Parasail |
| Output | $0.05 | |
| Cache read | $0.02 | −50% vs input |
| Context · max output | 131K · 16K | tokens |
Where to buy3 providers · per 1M tokens
| Provider | Input | Output | Context |
|---|---|---|---|
| AWS Bedrock | $0.04 | $0.08 | — |
| Parasail | $0.04 | $0.05 | 131K |
| DeepInfra | $0.05 | $0.10 | 131K |
In Google's line-upper 1M in / out
| Gemma 2 9B It | $0.03 / $0.06 | −12% |
| Gemma 4 E4B | $0.02 / $0.10 | −6% |
| Gemma 3 4B | $0.04 / $0.05 | this model |
| Gemma 3 1B IT | $0.04 / $0.08 | +18% |
| Gemma 3 12B | $0.05 / $0.10 | +47% |
| Gemma 1.1 7B It | $0.07 / $0.07 | +65% |
| DiffusionGemma 26B A4B | $0.05 / $0.15 | +76% |
Cheaper for the same workvs Gemma 3 4B
| Gemma 2 9B ItOpen | -12% | score wins 48% | $0.03 / $0.06 |
| GPT OSS 20B unstatedOpen | -15% | score wins 45% | $0.018 / $0.09 |
Pricing questions
How much does Gemma 3 4B cost?›
$0.04 per 1M input tokens and $0.05 per 1M output tokens. Cached input is $0.02.
Is Gemma 3 4B free to use?›
There is no free API tier, but the weights are open: you can run it on your own hardware for the cost of compute.
How much does prompt caching save?›
Cache reads cost $0.02 per 1M, 50% below the input price.
What is a cheaper alternative to Gemma 3 4B?›
Gemma 2 9B It costs -12% per task with a score wins 48% on Arena Text · overall.