$ cat models/mimo-v2.6-flash
MiMo-V2.6-Flash
Xiaomi
Check that an endpoint selling MiMo-V2.6-Flash really serves it. A test compares the endpoint's answers with answers collected from Xiaomi, through OpenRouter pinned to Xiaomi.
- Official Provider
- Xiaomi, through OpenRouter pinned to Xiaomi
- Reference Answers
- 400
- Declared Precision
- fp8
- Official Release
- fp8 e4m3, 128x128 blocks
- Checks
- Identity · Refusals (5 probes)
- Reference Data Updated
- 2026-10-16
How Well a Test Works
Size
Requests
Genuine Confirmed
Wrongly Flagged
Substitutes Caught
Standard
80
98.5%
0.5%
98.1% · hardest MiMo-V2.5
Thorough
160
99.5%
0.5%
100.0%
Simulated from the reference answers. A substitute that isn't caught ends inconclusive: it is never confirmed as this model.
- No competence check yet: the competence check needs hard tasks collected from the official provider, which this model doesn't have.