Skip to content
VECTOR WIREAI INTELLIGENCE
UTC

Inkling vs MiMo V2.5

27 SHARED BENCHMARKS

Across 27 shared benchmarks, Inkling scores higher on 17 and MiMo V2.5 on 10. The widest gap is τ-Bench V3 · Banking, where Inkling scores 29.1 against 8.7. MiMo V2.5 is the cheaper of the two on tracked API pricing ($0.14 against $1.00 per million input tokens).

THINKING MACHINESVSXIAOMI27 SHARED17–10 HEAD-TO-HEADNEWEST SCORE ADDED

At a glance

MakerThinking MachinesXiaomi
Released
Price per 1M tokens input / output$1.00 / $4.05$0.14 / $0.28
Cost of 1M in + 1M out$5.05$0.42 12.0× less
Head-to-head of 27 shared benchmarks17 wins10 wins
Scores tracked independently verified80 17 ◆51 3 ◆

Release dates: the vendor's own announcement for Inkling; Artificial Analysis for MiMo V2.5. Prices: Artificial Analysis for Inkling; Artificial Analysis for MiMo V2.5. ◆ = independently verified score.

Where each leadsby capability area · benchmark wins

AreaInklingWINSMiMo V2.5
Reasoning30Inkling leads 3 of 3 · widest: Humanity's Last Exam 31.9 vs 27.2
Coding22Even, 2–2 of 4
Agentic30Inkling leads 3 of 3 · widest: τ-Bench V3 · Banking 29.1 vs 8.7
Factuality21Inkling leads 2 of 3 · widest: SimpleQA Verified 43.9 vs 16.1
Instruction Following10Inkling leads 1 of 1 · widest: IFBench 79.8 vs 67.1
Long Context10Inkling leads 1 of 1 · widest: AA-LCR 77.3 vs 73
Math20Inkling leads 2 of 2 · widest: HMMT Feb. 2026 86.3 vs 82.6
Multimodal02MiMo V2.5 leads 2 of 2

Biggest gaps

Inkling pulls furthest ahead on

  1. τ-Bench V3 · Banking29.1 vs 8.7
  2. SimpleQA Verified43.9 vs 16.1
  3. AA-Omniscience · Accuracy41.5 vs 16.8

MiMo V2.5 pulls furthest ahead on

  1. AA-Omniscience · Non-hallucination68.1 vs 32.3
  2. Terminal-Bench 2.163.7 vs 55.1
  3. LMArena · WebDev1438 vs 1411

Every shared benchmark27 · grouped by area

Reasoning 3

BenchmarkInklingMARGINMiMo V2.5
CritPt5.43.7
GPQA Diamond87.284.9

Coding 4

BenchmarkInklingMARGINMiMo V2.5
LMArena · WebDev1411 ◆◆ 1438
SciCode4743.9

Agentic 3

BenchmarkInklingMARGINMiMo V2.5
GDPVal28.224.3
Terminal-Bench 4.010

Instruction Following 1

BenchmarkInklingMARGINMiMo V2.5
IFBench79.867.1

Long Context 1

BenchmarkInklingMARGINMiMo V2.5
AA-LCR77.373

Math 2

BenchmarkInklingMARGINMiMo V2.5
AIME 202697.193.6

Multimodal 2

BenchmarkInklingMARGINMiMo V2.5
MMMU-Pro73.575.4

Other shared benchmarks 8

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.