Sovereign inference — Gemma 3 12B
Your sovereign, private Gemma 3 12B
A reasoning model in the small tier, served on ToothFairyAI's AU, EU and US endpoints — private by default, billed in Units of Intelligence at the published rates below. No training on your data, no vendor lock-in.
AU · EU · US resident0.17 UoI/1M input0.51 UoI/1M output131K context
How Gemma 3 12B compares
Independent intelligence index for Gemma 3 12B against every peer in the Small tier (and the tier above where needed) — same source, same tests.
Latency — end-to-end seconds for a 500-token answer
Capability indexes
Domain-weighted agentic performance for Gemma 3 12B.
Capability index vs small peers
Finance & AccountingStrategy & OpsLegalEngineeringEconomics
The evaluations behind the index
Independent benchmark results for Gemma 3 12B — Elo scores as published; pass rates shown as percentages.
Per-evaluation economics
| Evaluation | Score | Time / task | Output tokens / task |
|---|---|---|---|
| AA-Briefcase | 0.0% | — | 67868 |
| GDPval-AA | -42436.0% | — | 80387 |
| AutomationBench-AA | 0.2% | — | 138 |
| Terminal-Bench 4.0 | 0.0% | — | 605 |
| SciCode | 16.4% | — | 723 |
| Humanity's Last Exam | 4.2% | — | 738 |
| GDP.pdf | 1.0% | — | 522 |
| CritPt | 0.0% | — | 2461 |
| AA-Omniscience | -7706.7% | — | 6 |
| AA-LCR (long context) | 8.3% | — | 110 |
Run Gemma 3 12B privately
One key, one endpoint, three jurisdictions.
- Size tierSmall
- Input rate0.17 UoI/1M
- Output rate0.51 UoI/1M
- Cached input rate—
- Context window131K tokens
- Max output8K tokens
- ReasoningYes — emits thinking traces
- Vision & video inputSupports image input
- Tool callingSupported
- Data residencyAU, EU and US endpoints — processing stays inside the region you call
Copyright 2026 ToothFairyAI PTY LTD. All Rights Reserved.
