AIEvals

Models

MiMo V2.6 Pro

Xiaomi’s MiMo V2.6 Pro release, with independently published coding and agentic results.

XiaomiReleased 21 September 2026Context 1M tokensOutput 128k tokensOpen weights$0.43 in / $0.87 out per Mtok

Family MiMo V2.632 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 AI55.20%12of 3312of 41Gemini 4 Argon13.70 pts
Artificial Analysis Intelligence IndexArtificial Analysis4612of 42no list readClaude Opus 5.512 pts
Terminal-Bench 4.0Artificial Analysis34.85%14of 40no list readClaude Sonnet 5.528.79 pts
Vibe Code Bench v1.1Vals AI85.22%12of 3812of 106Claude Sonnet 5.57.17 pts
Terminal-Bench Science 0.1Artificial Analysis5.71%18of 24no list readGPT-6 Astra57.62 pts
ProgramBenchVals AI0.50%13of 2913of 54Claude Opus 5.518.00 pts
MLCR-AAArtificial Analysis18.33%23of 37no list readClaude Sonnet 5.556.67 pts
AutomationBench-AAArtificial Analysis58.62%18of 40no list readGemini 4 Argon18.89 pts
SciCodeArtificial Analysis60.88%6of 40no list readClaude Opus 5.56.02 pts
GDP.pdfArtificial Analysis19.20%24of 40no list readGPT-6.1 Sol11.80 pts
CritPtArtificial Analysis26.57%13of 42no list readGPT-5.6 Sol5.72 pts
AA-OmniscienceArtificial Analysis8.3828of 42no list readClaude Opus 5.538.04
AA-LCR v1.1Artificial Analysis86.33%3of 42no list readKimi K32.34 pts
Terminal-Bench 2.1Vals AI67.79%27of 3532of 75Claude Opus 5.519.85 pts
ProofBench v1.1Vals AI70.00%15of 3417of 46Claude Sonnet 5.530.00 pts
Harvey's Legal Agent BenchmarkVals AI10.83%9of 3712of 73Muse Spark 1.214.59 pts
MysteryMechanismVals AI15.31%20of 2120of 21GPT-6 Astra37.84 pts
Code MigrationVals AI43.01%23of 3723of 71Claude Sonnet 5.526.82 pts
CyberBench v1.1Vals AI72.86%9of 3111of 43GPT-6 Sol5.12 pts
EMBVals AI62.86%24of 3727of 69Claude Fable 5.113.81 pts
Finance Agent (v2)Vals AI57.34%11of 3713of 73Gemini 4 Argon8.06 pts
IOIVals AI39.33%29of 3131of 38Gemini 4 Argon60.67 pts
Legal Research BenchVals AI47.12%12of 3712of 72Muse Spark 1.3 Max8.17 pts
MedCodeVals AI44.97%21of 3537of 104Claude Opus 518.60 pts
MedScribeVals AI88.31%10of 3611of 106Claude Opus 5.53.12 pts
Public Benefits Bench v1.1Vals AI68.94%6of 296of 45Claude Opus 57.99 pts
SAGEVals AI45.05%23of 3240of 90Claude Opus 4.711.05 pts
SRE BenchVals AI3.05%12of 1512of 16GPT-6 Astra53.82 pts
Tax Agent BenchVals AI64.94%16of 3718of 64Claude Fable 5.112.70 pts
GDPval-AA v2.1Artificial Analysis1,678.296of 42no list readClaude Opus 5.5167.88
AA-Briefcase v1.1Artificial Analysis1,519.5414of 41no list readClaude Opus 5.5302.31
Humanity's Last ExamArtificial Analysis49.35%10of 42no list readClaude Opus 5.512.00 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%Terminal-Bench Science 0.1%ProgramBench%
GPT-5.6 SolOpenAI58.01%4739.90%80.50%22.38%1.50%
GPT-6 SolOpenAI57.54%4843.94%87.82%30.00%2.00%
MiMo V2.6 ProXiaomi55.20%4634.85%85.22%5.71%0.50%
Claude Opus 4.8Anthropic55.10%4221.72%82.72%–1.00%
Grok 4.7xAI54.95%4625.76%86.17%–0.50%
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-Pro$0.1340.8 tok/s

Results by benchmark

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

BenchmarkIndependent runAs reported by a publisher
MLCR-AAMedical reasoning18.33%Artificial Analysisnot published
AutomationBench-AAWorkflow automation58.62%Artificial Analysisnot published
SciCodeScientific coding60.88%Artificial Analysisnot published
GDP.pdfDocument reasoning19.20%Artificial Analysisnot published
CritPtScientific reasoning26.57%Artificial Analysisnot published
AA-OmniscienceFactual knowledge8.38Artificial Analysisnot published
AA-LCR v1.1Long-context reasoning86.33%Artificial Analysisnot published
Terminal-Bench 4.0Agentic coding34.85%Artificial Analysis
31.31%Vals AI
not published
Terminal-Bench 2.1Agentic coding67.79%Vals AInot published
Vibe Code Bench v1.1Product coding85.22%Vals AInot published
Terminal-Bench Science 0.1Agentic science5.71%Artificial Analysis
2.86%Vals AI
not published
ProgramBenchAgentic coding0.50%Vals AInot published
ProofBench v1.1Mathematics70.00%Vals AInot published
Harvey's Legal Agent BenchmarkProfessional knowledge10.83%Vals AInot published
MysteryMechanismScientific reasoning15.31%Vals AInot published
Vals IndexComposite55.20%Vals AInot published
Code MigrationAgentic coding43.01%Vals AInot published
CyberBench v1.1Security72.86%Vals AInot published
EMBFinancial modelling62.86%Vals AInot published
Finance Agent (v2)Financial analysis57.34%Vals AInot published
IOICompetitive programming39.33%Vals AInot published
Legal Research BenchProfessional knowledge47.12%Vals AInot published
MedCodeHealthcare administration44.97%Vals AInot published
MedScribeHealthcare administration88.31%Vals AInot published
Public Benefits Bench v1.1Public services68.94%Vals AInot published
SAGEEducation45.05%Vals AInot published
SRE BenchSecurity3.05%Vals AInot published
Tax Agent BenchProfessional knowledge64.94%Vals AInot published
GDPval-AA v2.1Knowledge work1,678.29Artificial Analysisnot published
AA-Briefcase v1.1Knowledge work1,519.54Artificial Analysisnot published
Humanity's Last ExamMultidisciplinary reasoning49.35%Artificial Analysisnot published
Artificial Analysis Intelligence IndexComposite46Artificial 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.