Models/MiMo-V2-Flash

MDL

MiMo-V2-Flash

Xiaomi

Verified against Artificial Analysis · Jul 13, 2026

Capability

percentile index

Price /M

$0.1 / $0.3

in / out

Context

262.1K

tokens

Avg score

66.3

10 benchmarks

AA Intelligence

31.2

index

Weights

Closed

Released

Dec 16, 2025

About

MiMo-V2-Flash is a proprietary model in the MiMo family from Xiaomi. Benchmarked on 10 evals, averaging 66.3.

Benchmark scores · 10

avg 66.3 — higher is better

AIMEartificial-analysis · 2026-07

96.3

τ²-benchartificial-analysis · 2026-07

LiveCodeBenchartificial-analysis · 2026-07

86.8

GPQA Diamondartificial-analysis · 2026-07

84.6

MMLU-Proartificial-analysis · 2026-07

84.3

IFBenchartificial-analysis · 2026-07

64.2

AA Long-Context Reasoningartificial-analysis · 2026-07

SciCodeartificial-analysis · 2026-07

39.4

Terminal-Bench Hardartificial-analysis · 2026-07

Humanity's Last Examartificial-analysis · 2026-07

21.1

The research behind this model

all papers mentioning it →

MOPD: Multi-Teacher On-Policy Distillation for Capability Integration in LLM Post-Training▲ 15 on HF · Jun 29, 2026 Retrying vs Resampling in AI ControlMay 25, 2026 Positional Failures in Long-Context LLMs: A Blind Spot in Reasoning BenchmarksMay 22, 2026

Mentions matched by name in title/abstract — from the arXiv + HF daily corpus.

Details

PublisherXiaomi

FamilyMiMo

Parameters—

License—

WeightsClosed

ReleasedDec 16, 2025

Performance & price

Price in / out$0.1 / $0.3 per 1M

Blended price$0.15 / 1M

Math index96.3

Median across API providers, via Artificial Analysis. Blended = 3:1 input:output. For reasoning models, latency includes thinking time. Methodology →

History

full ledger →

Capability over time

Append-only ledger — every observed change to this model's numbers.

API providers

Artificial Analysis$0.3/M