AIEvals

Models

MiMo-V2.6-Flash

Specifications and API prices for MiMo-V2.6-Flash, recorded by Artificial Analysis.

XiaomiReleased 21 September 2026Context 1M tokensOutput Not published in the checked sourceOpen weights$0.14 in / $0.28 out per Mtok

Family MiMo-V2.6-Flash33 benchmarks with a published resultNo first place on the independent columns among the 46 models this site listsRead 17 October 2026

Where it stands

Rank compares tracked models with a score. Published shows the source’s rank and field size, when available.

BenchmarkScoreRankPublishedLeaderBehind
Vals IndexVals AI53.23%17of 3317of 41Gemini 4 Argon15.67 pts
Artificial Analysis Intelligence IndexArtificial Analysis3833of 42no list readClaude Opus 5.520 pts
Terminal-Bench 4.0Artificial Analysis22.73%22of 40no list readClaude Sonnet 5.540.91 pts
Vibe Code Bench v1.1Vals AI78.96%21of 3822of 106Claude Sonnet 5.513.43 pts
ProgramBenchVals AI0.50%13of 2913of 54Claude Opus 5.518.00 pts
AutomationBench-AAArtificial Analysis64.05%9of 40no list readGemini 4 Argon13.46 pts
SciCodeArtificial Analysis51.27%35of 40no list readClaude Opus 5.515.63 pts
GDP.pdfArtificial Analysis9.00%39of 40no list readGPT-6.1 Sol22.00 pts
CritPtArtificial Analysis12.00%37of 42no list readGPT-5.6 Sol20.29 pts
AA-OmniscienceArtificial Analysis-12.6841of 42no list readClaude Opus 5.559.1
AA-LCR v1.1Artificial Analysis74.33%41of 42no list readKimi K314.34 pts
Terminal-Bench 2.1Vals AI76.40%16of 3516of 75Claude Opus 5.511.24 pts
ProofBench v1.1Vals AI63.00%17of 3419of 46Claude Sonnet 5.537.00 pts
Harvey's Legal Agent BenchmarkVals AI11.25%7of 379of 73Muse Spark 1.214.17 pts
MysteryMechanismVals AI21.62%16of 2116of 21GPT-6 Astra31.53 pts
BioMysteryBenchVals AI69.26%11of 1911of 21Claude Sonnet 5.511.85 pts
Code MigrationVals AI40.93%26of 3726of 71Claude Sonnet 5.528.90 pts
CyberBench v1.1Vals AI75.36%5of 315of 43GPT-6 Sol2.62 pts
EMBVals AI65.46%21of 3721of 69Claude Fable 5.111.21 pts
Finance Agent (v2)Vals AI56.28%13of 3717of 73Gemini 4 Argon9.12 pts
IOIVals AI47.67%24of 3125of 38Gemini 4 Argon52.33 pts
Legal Research BenchVals AI37.98%28of 3731of 72Muse Spark 1.3 Max17.31 pts
MedCodeVals AI41.06%33of 3556of 104Claude Opus 522.51 pts
MedScribeVals AI85.28%22of 3628of 106Claude Opus 5.56.15 pts
Public Benefits Bench v1.1Vals AI67.59%12of 2912of 45Claude Opus 59.34 pts
SAGEVals AI43.53%26of 3248of 90Claude Opus 4.712.57 pts
SRE BenchVals AI4.20%10of 1510of 16GPT-6 Astra52.67 pts
Tax Agent BenchVals AI59.90%28of 3733of 64Claude Fable 5.117.74 pts
GDPval-AA v2.1Artificial Analysis1,600.0615of 42no list readClaude Opus 5.5246.11
AA-Briefcase v1.1Artificial Analysis1,498.5917of 41no list readClaude Opus 5.5323.26
Humanity's Last ExamArtificial Analysis35.08%40of 42no list readClaude Opus 5.526.27 pts

Nearby on Vals Index

The two models above and below on the Vals Index (Vals AI), on every column where this model has a published number.

ModelVals Index%Artificial Analysis Intelligence IndexindexTerminal-Bench 4.0%Vibe Code Bench v1.1%ProgramBench%
Gemini 3.8 FlashGoogle54.83%4119.70%78.65%1.00%
GLM 5.3Z.ai53.51%4541.92%78.13%1.50%
MiMo-V2.6-FlashXiaomi53.23%3822.73%78.96%0.50%
GPT-5.6 TerraOpenAI53.09%4235.35%74.59%0.50%
Grok 4.6xAI52.09%4417.17%76.24%–
Bold marks a figure better than this model’s on the same column, whichever model it belongs to. Ranks cover the models shown from the stated source.

Cost and speed, as published

Additional measurements from Artificial Analysis’ own row. Read 30 September 2026.

As publishedCost per taskOutput speed
MiMo-V2.6-Flash$0.0656.6 tok/s

Results by benchmark

Independent and vendor results retain their own harnesses and effort settings.

BenchmarkIndependent runAs reported by a publisher
AutomationBench-AAWorkflow automation64.05%Artificial Analysisnot published
SciCodeScientific coding51.27%Artificial Analysisnot published
GDP.pdfDocument reasoning9.00%Artificial Analysisnot published
CritPtScientific reasoning12.00%Artificial Analysisnot published
AA-OmniscienceFactual knowledge-12.68Artificial Analysisnot published
AA-LCR v1.1Long-context reasoning74.33%Artificial Analysisnot published
Terminal-Bench 4.0Agentic coding22.73%Artificial Analysis
24.24%Vals AI
not published
Terminal-Bench 2.1Agentic coding76.40%Vals AInot published
MMMU ProMultimodal reasoning
73.06%Artificial Analysis
not published
Vibe Code Bench v1.1Product coding78.96%Vals AInot published
Terminal-Bench Science 0.1Agentic science
4.29%Vals AI
not published
ProgramBenchAgentic coding0.50%Vals AInot published
ProofBench v1.1Mathematics63.00%Vals AInot published
Harvey's Legal Agent BenchmarkProfessional knowledge11.25%Vals AInot published
MysteryMechanismScientific reasoning21.62%Vals AInot published
Vals IndexComposite53.23%Vals AInot published
BioMysteryBenchAgentic science69.26%Vals AInot published
Code MigrationAgentic coding40.93%Vals AInot published
CyberBench v1.1Security75.36%Vals AInot published
EMBFinancial modelling65.46%Vals AInot published
Finance Agent (v2)Financial analysis56.28%Vals AInot published
IOICompetitive programming47.67%Vals AInot published
Legal Research BenchProfessional knowledge37.98%Vals AInot published
MedCodeHealthcare administration41.06%Vals AInot published
MedScribeHealthcare administration85.28%Vals AInot published
Public Benefits Bench v1.1Public services67.59%Vals AInot published
SAGEEducation43.53%Vals AInot published
SRE BenchSecurity4.20%Vals AInot published
Tax Agent BenchProfessional knowledge59.90%Vals AInot published
GDPval-AA v2.1Knowledge work1,600.06Artificial Analysisnot published
AA-Briefcase v1.1Knowledge work1,498.59Artificial Analysisnot published
Humanity's Last ExamMultidisciplinary reasoning35.08%Artificial Analysisnot published
Artificial Analysis Intelligence IndexComposite38Artificial Analysisnot published
Independent column: the suite named under each figure, one harness per suite. Publisher column: each lab’s own table, reproduced on the benchmark’s page. Publisher ranks compare reported scores in the same table row only; ties share a rank and unreported cells are excluded.

Sources

Read on 17 October 2026.