Sovereign inference — GPT OSS 20B
Your sovereign, private GPT OSS 20B
A reasoning model in the medium 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 resident0.43 UoI/1M input1.51 UoI/1M output131K context
How GPT OSS 20B compares
Independent intelligence index for GPT OSS 20B against every peer in the Medium 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 GPT OSS 20B.
Capability index vs medium peers
Finance & AccountingStrategy & OpsLegalHealthcare & MedicalEngineeringEconomics
The evaluations behind the index
Independent benchmark results for GPT OSS 20B — Elo scores as published; pass rates shown as percentages.
Per-evaluation economics
| Evaluation | Score | Time / task | Output tokens / task |
|---|---|---|---|
| AA-Briefcase | 0.0% | 2.2 min | 25109 |
| GDPval-AA | 326.46 | 3.3 min | 37012 |
| AutomationBench-AA | 0.2% | 0.4 min | 4861 |
| Terminal-Bench 4.0 | 0.0% | 2.2 min | 24838 |
| SciCode | 38.9% | 0.9 min | 9747 |
| Humanity's Last Exam | 11.0% | 1.4 min | 16181 |
| GDP.pdf | 2.0% | 1.0 min | 10833 |
| CritPt | 1.4% | 3.4 min | 37604 |
| AA-Omniscience | -6305.0% | 0.2 min | 2129 |
| AA-LCR (long context) | 34.7% | 0.3 min | 3171 |
Run GPT OSS 20B privately
One key, one endpoint, three jurisdictions.
- Size tierMedium
- Input rate0.43 UoI/1M
- Output rate1.51 UoI/1M
- Cached input rate—
- Context window131K tokens
- Max output66K tokens
- ReasoningYes — emits thinking traces
- Vision & video inputText only
- Tool callingSupported
- Data residencyAU, EU and US endpoints — processing stays inside the region you call
Copyright 2026 ToothFairyAI PTY LTD. All Rights Reserved.
