VECTOR WIREAI INTELLIGENCE
UTC
Refresh Models Deals Regulatory Sources

GPT-5.6 Luna vs MiMo V2.6 Pro

16 SHARED BENCHMARKS

Across 16 shared benchmarks, GPT-5.6 Luna scores higher on 1 and MiMo V2.6 Pro on 15. The widest gap is Humanity's Last Exam, where MiMo V2.6 Pro scores 49.4 against 7.2. Tracked API pricing per million tokens: GPT-5.6 Luna $0.20 in / $1.20 out, MiMo V2.6 Pro $0.43 in / $0.87 out.

OPENAIVSXIAOMI16 SHARED115 HEAD-TO-HEADNEWEST SCORE ADDED

At a glance

MakerOpenAIXiaomi
Released
Price per 1M tokens input / output$0.20 / $1.20$0.43 / $0.87
Cost of 1M in + 1M out$1.40$1.30 1.1× less
Head-to-head of 16 shared benchmarks1 wins15 wins
Scores tracked independently verified183 4321 1

Release dates per Artificial Analysis. Prices: Artificial Analysis for GPT-5.6 Luna; Artificial Analysis for MiMo V2.6 Pro. ◆ = independently verified score.

Where each leadsby capability area · benchmark wins

AreaGPT-5.6 LunaWINSMiMo V2.6 Pro
Reasoning02MiMo V2.6 Pro leads 2 of 2 · widest: Humanity's Last Exam 7.2 vs 49.4
Coding02MiMo V2.6 Pro leads 2 of 2 · widest: Terminal-Bench 2.1 39 vs 89.9
Agentic03MiMo V2.6 Pro leads 3 of 3 · widest: GDPVal 20.3 vs 58.7
Factuality02MiMo V2.6 Pro leads 2 of 2 · widest: AA-Omniscience · Non-hallucination 25 vs 59.4
Long Context01MiMo V2.6 Pro leads 1 of 1 · widest: AA-LCR 42.7 vs 86.3

Biggest gaps

GPT-5.6 Luna pulls furthest ahead on

No ratified-area lead of 3 points or more.

MiMo V2.6 Pro pulls furthest ahead on

  1. Humanity's Last Exam49.4 vs 7.2
  2. GDPVal58.7 vs 20.3
  3. AA-Omniscience · Non-hallucination59.4 vs 25

Every shared benchmark16 · grouped by area

Reasoning 2

Coding 2

Agentic 3

GDPVal20.358.7
Terminal-Bench 4.0134.8

Long Context 1

AA-LCR42.786.3

Other shared benchmarks 6

Terminal-Bench 4.017.334.9

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, direct), otherwise the lowest tracked offer.