Your sovereign, private Qwen 3.8 Flash
Qwen 3.8 Flash is the fast variant of the Qwen3.8 family: a lightweight reasoner with a 1M-token context, native tool calling and strong coding-agentic performance at very low cost. A reasoning model in the small tier, served on ToothFairyAI's AU, EU and US endpoints — private by default, billed in US dollars per million tokens at the published rates below. No training on your data, no vendor lock-in.
How Qwen 3.8 Flash compares
Independent intelligence index for Qwen 3.8 Flash against every peer in the Small tier (and the tier above where needed) — same source, same tests.
Throughput — tokens per second, median
Latency — end-to-end seconds for a 500-token answer
Capability indexes
Domain-weighted agentic performance for Qwen 3.8 Flash.
Capability index vs small peers
The evaluations behind the index
Independent benchmark results for Qwen 3.8 Flash — Elo scores as published; pass rates shown as percentages.
Per-evaluation economics
| Evaluation | Score | Time / task | Output tokens / task |
|---|---|---|---|
| AA-Briefcase | 1579.93 | 80.6 min | 311692 |
| GDPval-AA | 1627.86 | 39.8 min | 153799 |
| AutomationBench-AA | 55.9% | 5.5 min | 21459 |
| Terminal-Bench 4.0 | 25.3% | 60.9 min | 235646 |
| SciCode | 50.6% | 6.6 min | 25479 |
| Humanity's Last Exam | 38.0% | 13.2 min | 50969 |
| GDP.pdf | 15.6% | 2.6 min | 10174 |
| CritPt | 11.1% | 30.1 min | 116306 |
| AA-Omniscience | -971.7% | 1.1 min | 4380 |
| AA-LCR (long context) | 79.7% | 0.8 min | 3274 |
Run Qwen 3.8 Flash privately
One key, one endpoint, three jurisdictions.
- Size tierSmall
- Serving routeStandard
- Input rate$0.19/1M
- Output rate$0.59/1M
- Cached input rate—
- Context window1000K tokens
- Max output64K tokens
- ReasoningYes — emits thinking traces
- Vision & video inputText only
- Tool callingSupported
- Data residencyAU, EU and US endpoints — processing stays inside the region you call