Skip to main content
XI

MiMo V2.6 Flash

xiaomi/mimo-v2.6-flash

Xiaomi multimodal reasoning model for high-frequency professional workloads. It accepts text, image, video and audio input, generates text, and supports a one-million-token context window, 128K maximum output, tool use, structured output and prompt caching.

Compare
Context window
1M
Input
Audio · Image · Text · Video
Output
Text
Released
—
Max output
128K
Capabilities
Attachments, Reasoning, Structured, Tool call

Pricing

IN

$0.14

USD / 1,000,000 token

Cache read

$0.0028

USD / 1,000,000 token

OUT

$0.28

USD / 1,000,000 token

Performance

Overall capability

53.6/100

Coding
52.1
Reasoning
46.8
Professional work
63.9
Knowledge & language
48.0
MysteryMechanismReasoningVals AI2026-11-05
ProgramBenchCodingVals AI2026-11-05
Finance AgentProfessional workVals AI2026-11-05
Terminal-Bench 4.0 (AA rerun)MiMo-V2.6-FlashCodingArtificial Analysis2026-11-05
AA-BriefcaseProfessional workArtificial Analysis2026-11-05
AutomationBench-AAMiMo-V2.6-FlashProfessional workArtificial Analysis2026-11-05
AA-OmniscienceMiMo-V2.6-FlashKnowledge & languageArtificial Analysis2026-11-05
Harvey Legal Agent BenchmarkProfessional workVals AI / Harvey2026-11-05

Providers

No providers found

ModelRegistry · Catalog updated 2026-11-05