Google Gemma 3

by GoogleChat

Open Gemma instruction model for efficient chat and self-hosted deployments

Context
125K tokens
Max output
4,096 tokens
Released
2025-01-01
Weights
Open

google/gemma-3

Specifications

Full facts as recorded, incl. what is unknown.

Released
2025-01-01
Weights
Open
Input
Text, Image
Output
Text
Context window
125,000 max across endpoints
Max output
4,096
Providers
1
Epoch Capabilities Index
not evaluated
Direct price range
$0.15 in · $0.30 out per 1M tokens · from its one provider
More details
Last updated
2025-01-01
Knowledge cutoff
2024-12

Capabilities

Across all endpoints: supported if any endpoint supports it.

  • Tool calling
  • Parallel toolsunknown
  • Structured outputunknown
  • Reasoning
  • Image input
  • PDF input
  • Audio input
  • Audio outputunknown
  • Video input
  • Prompt cachingunknown
  • Web searchunknown
  • Computer useunknown
  • System messagesunknown
  • Assistant prefillunknown
  • Temperature
  • Streamingunknown
  • Attachments

Endpoints

One row per provider and variant; gateways are listed separately.

Loading endpoints…

Price history

No change since 19 Oct 2026. The cheapest direct price has been $0.15 in / $0.30 out per 1M tokens at Inference on every data refresh since then.

Tracked from 19 Oct 2026. Headline-eligible direct prices only, so a price the sanity check flags never draws a spike. Series as JSON

Cost calculator

Loading endpoints…

More from Google

Newest Google models in the catalogue. All Google models

ModelReleasedContextListed price
Gemini 4 Argon2026-10-131.05Mfrom $2.00
Gemma 4 31B Scarlet2026-10-08262K$0.10 via NanoGPT
Gemma 4 26B A4B Chimera X Heretic2026-10-08262K$0.12 via NanoGPT
Gemini Nano Banana 2.12026-10-06131Kfrom $1.50
Gemini 3.8 Flash TTS2026-09-238.2Kunpriced
Gemini 3.8 Flash-Lite TTS2026-09-238.2Kunpriced
DiffusionGemma2026-09-19262K$0.05 via NanoGPT
Gemma 4 26B A4B Cybersecurity2026-09-19262K$0.106 via NanoGPT
Heads up: prices are indicative list rates scraped from provider pricing pages; verify with the provider before relying on them. Report incorrect data for this model