RentGPU RentGPU
Tìm GPU →
by Google DeepMind

Google: Gemma 3 12B.

multimodal closed laptop+ 12B params 131K ctx
Cheapest input
$0.04/M
on DeepInfra
Cheapest output
$0.13/M
on DeepInfra
Fastest
53 tok/s
on OpenRouter
Hosted equiv.
~$0.05/hr
@ 100 tok/s on DeepInfra

Gemma 3 introduces multimodality, supporting vision-language input and text outputs. It handles context windows up to 128k tokens, understands over 140 languages, and offers improved math, reasoning, and chat capabilitie

Hosted API only

No self-host path — closed weights.

Google: Gemma 3 12B's weights aren't published. Use it via the access providers below.

Where to use it

Cheapest hosted endpoints.

Provider Access $/M in $/M out
DeepInfra hosted inference $0.04 $0.13 Launch ↗
OpenRouter api aggregator $0.04 $0.13 Launch ↗
Performance

Speed across providers.

Tokens/sec and time-to-first-token measured against the same prompt template on each provider's API.

Provider Tokens/sec TTFT Total
OpenRouter 53.2 506 ms 1598 ms
FAQ

Frequently asked.

How do I run Google: Gemma 3 12B?
Google: Gemma 3 12B is a closed-source API model. The cheapest way to access it is through the API providers listed on this page (direct API, aggregators, and hosted chat UIs).
Where can I access Google: Gemma 3 12B?
Google: Gemma 3 12B is available via DeepInfra, OpenRouter. Each access option lists its own pricing (per million tokens or hourly hosting).
How much does it cost to run Google: Gemma 3 12B?
API pricing starts at $0.04/M input tokens and $0.13/M output tokens. Self-hosting cost depends on the GPU you rent — see the Run It Yourself tab.
Is Google: Gemma 3 12B open-source or proprietary?
Google: Gemma 3 12B is a proprietary model from Google DeepMind. Access is API-only — there are no public weights to download.