Free Gemma model via API
Updated:
Open Gemma models for experiments and light production. This page is separate from Gemini: it is a different model line.
Read more
Gemma does not inherit Google AI Studio Gemini quotas. Read the limit on the selected host.
26 models found
| № | Model | Power | Provider | Parameters | Context | Limit | |
|---|---|---|---|---|---|---|---|
| 1 |
Google: Gemma 4 26B A4B (free)
|
⚖️ | OpenRouter | 26B | 262K | Free tier | Open |
| 2 |
Google: Gemma 4 31B (free)
|
⚖️ | OpenRouter | 31B | 262K | Free tier | Open |
| 3 |
Gemma 3 27B It
|
⚖️ | Together AI | 27B | 66K | Fair Use (По мере нагрузки серверов) | Open |
| 4 |
Gemma 3 1b it
|
⚡ | Together AI | 1B | 33K | Fair Use (По мере нагрузки серверов) | Open |
| 5 |
Gemma 3 4b it
|
⚡ | Together AI | 4B | 66K | Fair Use (По мере нагрузки серверов) | Open |
| 6 |
Gemma 2B It
|
⚡ | Together AI | 2B | 8K | Fair Use (По мере нагрузки серверов) | Open |
| 7 |
Gemma 2 9B It
|
⚡ | Together AI | 9B | 8K | Fair Use (По мере нагрузки серверов) | Open |
| 8 |
Gemma 3 270M It
|
⚡ | Together AI | ? | 33K | Fair Use (По мере нагрузки серверов) | Open |
| 9 | Medgemma 27B Text It | ⚖️ | Together AI | 27B | 131K | Fair Use (По мере нагрузки серверов) | Open |
| 10 |
Gemma 4 E4B-it
|
⚡ | Together AI | ? | 131K | Fair Use (По мере нагрузки серверов) | Open |
| 11 |
Gemma 4 26B A4b It
|
⚖️ | Together AI | 26B | 262K | Fair Use (По мере нагрузки серверов) | Open |
| 12 |
Gemma 4 E2B-it
|
⚡ | Together AI | ? | 131K | Fair Use (По мере нагрузки серверов) | Open |
| 13 |
Gemma 3 270M It Lora
|
⚡ | Together AI | ? | 33K | Fair Use (По мере нагрузки серверов) | Open |
| 14 |
Gemma 3 27B It Lora
|
⚖️ | Together AI | 27B | ? | Fair Use (По мере нагрузки серверов) | Open |
| 15 |
Gemma 4 31B It Lora
|
⚖️ | Together AI | 31B | 262K | Fair Use (По мере нагрузки серверов) | Open |
| 16 |
Gemma 4 12B It
|
⚖️ | Together AI | 12B | 262K | Fair Use (По мере нагрузки серверов) | Open |
| 17 | Codegemma 1.1 7b | ⚡ | NVIDIA Build | 7B | ? | Free serverless inference for development; limits and model availability may change. | Open |
| 18 | Codegemma 7b | ⚡ | NVIDIA Build | 7B | ? | Free serverless inference for development; limits and model availability may change. | Open |
| 19 |
Gemma 2b
|
⚡ | NVIDIA Build | 2B | ? | Free serverless inference for development; limits and model availability may change. | Open |
| 20 |
Gemma 3 12b It
|
⚖️ | NVIDIA Build | 12B | ? | Free serverless inference for development; limits and model availability may change. | Open |
| 21 |
Gemma 3 4b It
|
⚡ | NVIDIA Build | 4B | ? | Free serverless inference for development; limits and model availability may change. | Open |
| 22 |
Gemma 4 31b It
|
⚖️ | NVIDIA Build | 31B | ? | Free serverless inference for development; limits and model availability may change. | Open |
| 23 | Recurrentgemma 2b | ⚡ | NVIDIA Build | 2B | ? | Free serverless inference for development; limits and model availability may change. | Open |
| 24 | Diffusiongemma 26b A4b It | ⚖️ | UnoRouter | 26B | 262K | No-card free models use shared capacity and per-model limits; availability can change. | Open |
| 25 |
Gemma 4 26b
|
⚖️ | UnoRouter | 26B | 262K | No-card free models use shared capacity and per-model limits; availability can change. | Open |
| 26 |
Gemma 4 31b It
|
⚖️ | UnoRouter | 31B | 262K | No-card free models use shared capacity and per-model limits; availability can change. | Open |