Same open weights, served by 2 providers. Weights: google/gemma-3-12b-it.
| Provider | Their model id | Serving | Input | Output | Cached in | Blended 3:1 | Source | |
|---|---|---|---|---|---|---|---|---|
DeepInfra | google/gemma-3-12b-it | flexFull precision0.8× standard | $0.04 | $0.12 | — | $0.06 | Verified 5h ago · Changed 7h ago | |
DeepInfra | google/gemma-3-12b-it | standardFull precision | $0.05 | $0.15 | — | $0.075 | Verified 5h ago · Changed 7h ago | |
DeepInfra | google/gemma-3-12b-it | priorityFull precision1.5× standard | $0.075 | $0.225 | — | $0.112 | Verified 5h ago · Changed 7h ago | |
AWS Bedrock | google.gemma-3-12b-it | standardQuantization not disclosed | $0.09 | $0.29 | — | $0.14 | Verified 7h ago · Changed 7h ago |
Same weights don't always mean identical outputs: quantization, context limits and serving stacks can change quality, especially for tool calling. Prices are USD per 1M tokens; hyperscaler prices are for the reference region shown.
| DeepInfra bfloat16 | $8/mo |
| AWS Bedrock | $14.8/mo |
Cache hits use each provider's cached-input price where it has one. Cache write costs aren't included.