- input
- $0.08
- cached
- $0.04
- output
- $0.35
- Context
- 131Ktokens
- Released
- 1y ago
- Size
- 27Bparameters
Cost/1M tokens
Open-weight builds
Ollama variants
Every published Ollama tag matched to this model, including parameter editions, instruction/base variants, and quantized builds.
| Variant | Parameters | Type | Quantization | Size | Context | Input |
|---|---|---|---|---|---|---|
| gemma3:270m-it-bf16 | 270M | — | BF16 | 543 MB | 32K | text |
| gemma3:270m-it-fp16 | 270M | — | FP16 | 543 MB | 32K | text |
| gemma3:270m | 270M | — | Q8_0 | 292 MB | 32K | text |
| gemma3:270m-it-qat | 270M | — | — | 241 MB | 32K | text |
| gemma3:1b-it-fp16 | 1B | — | FP16 | 2 GB | 32K | text |
| gemma3:1b | 1B | — | Q4_K_M | 806 MB | 32K | text |
| gemma3:1b-it-q8_0 | 1B | — | Q8_0 | 1.1 GB | 32K | text |
| gemma3:1b-it-qat | 1B | — | — | 1 GB | 32K | text |
| gemma3:4b-it-bf16 | 4B | — | BF16 | 8.6 GB | 128K | text + image |
| gemma3:latest | 4B | — | Q4_K_M | 3.4 GB | 128K | text + image |
| gemma3:4b-it-q8_0 | 4B | — | Q8_0 | 5 GB | 128K | text + image |
| gemma3:4b-it-qat | 4B | — | — | 4 GB | 128K | text + image |
| gemma3:12b-it-bf16 | 12B | — | BF16 | 24 GB | 128K | text + image |
| gemma3:12b | 12B | — | Q4_K_M | 8.2 GB | 128K | text + image |
| gemma3:12b-it-q8_0 | 12B | — | Q8_0 | 13 GB | 128K | text + image |
| gemma3:12b-it-qat | 12B | — | — | 8.9 GB | 128K | text + image |
| gemma3:27b-it-bf16 | 27B | — | BF16 | 55 GB | 128K | text + image |
| gemma3:27b | 27B | — | Q4_K_M | 17 GB | 128K | text + image |
| gemma3:27b-it-q8_0 | 27B | — | Q8_0 | 30 GB | 128K | text + image |
| gemma3:27b-it-qat | 27B | — | — | 18 GB | 128K | text + image |
Deployment graph
Can I run this?
Vertical Tabs only shows compatibility when it has structured evidence tying this exact model to hardware, runtime, quantization, memory use, and observed performance.