[mintok]
Get started →

Gemma 3 27B — pricing & specs

Gemma 3 27B is a fast model from Google, at capability tier 3 of 8. Live $/M-token pricing and specs from the Mintok catalog, as of 2026-05-26.

$/M-token pricing

as of 2026-05-26 · Google
Input
$0.09
per M tokens
Output
$0.33
per M tokens
Cached input
—
per M tokens

What Gemma 3 27B costs in practice

ScenarioEst. cost
1M input + 1M output tokens$0.420
A 10K-token request (8K in / 2K out)$0.0014
1,000 agent runs (~2K in / 500 out each)$0.345
1M-token RAG chat session (900K cached in / 100K out)$0.114

Estimated from list price (as of 2026-05-26) at the token mixes shown; your real cost depends on prompt/response length and cache hit-rate. Model your own volume →

Specifications

Vendor
Google
Family
Gemma 3
Capability tier
3 / 8
Size class
FAST
Category
GENERAL
Parameters
27B total · 27B active (MoE)
Context window
131,072 tokens
Licensing
Proprietary

Carbon footprint (modeled)

Energy
0.105
kWh / M-token
Emissions
50.6
gCO₂ / M-token

Modeled decode on a reference NVIDIA Rubin Ultra VR300 at 50% MBU, batch 32, global-avg grid. Compare all models & grids →

Alternatives at tier 3

Same capability tier, sorted by output price — the cheaper-capable options.

FAQ

How much does Gemma 3 27B cost per token?

Gemma 3 27B costs $0.09 per million input tokens and $0.33 per million output tokens, via Google, as of 2026-05-26.

What does a typical request with Gemma 3 27B cost?

At list price, a 10K-token request (8K in / 2K out) with Gemma 3 27B costs about $0.0014, and 1,000 agent runs (~2K in / 500 out each) about $0.345.

What is Gemma 3 27B's context window?

Gemma 3 27B supports a 131,072-token context window.

What capability tier is Gemma 3 27B?

Gemma 3 27B sits at capability tier 3 of 8 on the Mintok scale (FAST size class).

Is Gemma 3 27B open-source or proprietary?

Gemma 3 27B is a proprietary model from Google, accessed via API.

Put Gemma 3 27B to work

Benchmark it on your prompts, or size a deployment and compare $/M-token against alternatives.

Benchmark your prompts →Size a deployment →Cheapest fast models →More from Google →Cheapest by tier →

Pricing and specs are maintained by Mintok's catalog watchers from public sources. Not investment advice. Methodology.