MiMo V2.5 vs Qwen3.5 397B A17B
31 SHARED BENCHMARKSAcross 31 shared benchmarks, MiMo V2.5 scores higher on 14 and Qwen3.5 397B A17B on 16, with 1 level. The widest gap is AA-Omniscience · Non-hallucination, where MiMo V2.5 scores 68.1 against 11.1. MiMo V2.5 is the cheaper of the two on tracked API pricing ($0.14 against $0.60 per million input tokens).
At a glance
Release dates per Artificial Analysis. Prices: Artificial Analysis for MiMo V2.5; Alibaba's own price page for Qwen3.5 397B A17B. ◆ = independently verified score.
Where each leadsby capability area · benchmark wins
Biggest gaps
MiMo V2.5 pulls furthest ahead on
- AA-Omniscience · Non-hallucination68.1 vs 11.1
- GDPVal24.3 vs 14
- Terminal-Bench 2.163.7 vs 51.3
Qwen3.5 397B A17B pulls furthest ahead on
- AA-Omniscience · Accuracy30.8 vs 16.8
- SimpleQA Verified26 vs 16.1
- τ-Bench V3 · Banking13.4 vs 8.7
Every shared benchmark31 · grouped by area
Reasoning 3
Coding 6
Agentic 3
Factuality 3
Instruction Following 1
Long Context 1
Math 2
Multimodal 2
Other shared benchmarks 10
Each score is the one the model's own page shows — the most authoritative tracked result for that benchmark, on the benchmark's standard methodology (source-attributed). ↓ marks lower-is-better metrics; ◆ an independently verified score. Open either model for its full surface, provenance and pricing. Quoted rates are the price-setter row we currently track for each model — its direct or vendor-official listing where one exists (direct, alibaba-official), otherwise the lowest tracked offer.