AIEvals

Models

Qwen3.8 Max (0902)

Specifications and API prices for Qwen3.8 Max (0902), recorded by Artificial Analysis.

AlibabaReleased 2 September 2026Context 983,616 tokensOutput Not published in the checked sourcePrivate weights$2.00 in / $6.00 out per Mtok

Family Qwen3.822 benchmarks with a published result1 independent first place 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
Artificial Analysis Intelligence IndexArtificial Analysis4514of 42no list readClaude Opus 5.513 pts
Terminal-Bench 4.0Artificial Analysis38.89%12of 40no list readClaude Sonnet 5.524.75 pts
Terminal-Bench Science 0.1Artificial Analysis11.90%11of 24no list readGPT-6 Astra51.43 pts
APEX-Agents-AAArtificial Analysis42.40%1of 7no list readGemini 3 5 Flash4.65 pts
τ³-BankingArtificial Analysis47.84%5of 30no list readQwen3.8 Max3.50 pts
ITBench-AAArtificial Analysis40.25%15of 18no list readGPT-5.6 Sol15.96 pts
AA-AnalystAgentArtificial Analysis45.00%9of 18no list readGemini 3.7 Flash15.00 pts
MLCR-AAArtificial Analysis20.00%20of 37no list readClaude Sonnet 5.555.00 pts
Harvey LAB-AAArtificial Analysis93.65%3of 20no list readMuse Spark 1.3 Max1.83 pts
EnterpriseOps-Gym-AAArtificial Analysis47.63%5of 22no list readClaude Fable 53.49 pts
AutomationBench-AAArtificial Analysis56.19%25of 40no list readGemini 4 Argon21.32 pts
SciCodeArtificial Analysis52.08%32of 40no list readClaude Opus 5.514.82 pts
GDP.pdfArtificial Analysis22.80%13of 40no list readGPT-6.1 Sol8.20 pts
CritPtArtificial Analysis17.71%28of 42no list readGPT-5.6 Sol14.58 pts
AA-OmniscienceArtificial Analysis11.9826of 42no list readClaude Opus 5.534.44
AA-LCR v1.1Artificial Analysis80.33%26of 42no list readKimi K38.34 pts
GDPval-AA v2.1Artificial Analysis1,663.258of 42no list readClaude Opus 5.5182.92
AA-Briefcase v1.1Artificial Analysis1,624.466of 41no list readClaude Opus 5.5197.39
Humanity's Last ExamArtificial Analysis43.10%25of 42no list readClaude Opus 5.518.25 pts

Nearby on Artificial Analysis Intelligence Index

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

ModelArtificial Analysis Intelligence IndexindexTerminal-Bench 4.0%Terminal-Bench Science 0.1%
MiMo V2.6 ProXiaomi4634.85%5.71%
GLM 5.3Z.ai4541.92%9.52%
Qwen3.8 Max (0902)Alibaba4538.89%11.90%
Grok 4.6xAI4417.17%6.19%
Kimi K3Moonshot AI4412.63%–
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
Qwen3.8 Max (0902)$5.4139 tok/s

Results by benchmark

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

BenchmarkIndependent runAs reported by a publisher
APEX-Agents-AAKnowledge work42.40%Artificial Analysisnot published
τ³-BankingTool use47.84%Artificial Analysisnot published
ITBench-AAIT operations40.25%Artificial Analysisnot published
AA-AnalystAgentData analysis45.00%Artificial Analysisnot published
MLCR-AAMedical reasoning20.00%Artificial Analysisnot published
Harvey LAB-AALegal reasoning93.65%Artificial Analysisnot published
EnterpriseOps-Gym-AAEnterprise workflows47.63%Artificial Analysisnot published
AutomationBench-AAWorkflow automation56.19%Artificial Analysisnot published
SciCodeScientific coding52.08%Artificial Analysisnot published
GDP.pdfDocument reasoning22.80%Artificial Analysisnot published
CritPtScientific reasoning17.71%Artificial Analysisnot published
AA-OmniscienceFactual knowledge11.98Artificial Analysisnot published
AA-LCR v1.1Long-context reasoning80.33%Artificial Analysisnot published
Terminal-Bench 4.0Agentic coding38.89%Artificial Analysisnot published
Terminal-Bench 2.1Agentic coding
88.76%Artificial Analysis
not published
GPQA DiamondScience reasoning
92.83%Artificial Analysis
not published
MMMU ProMultimodal reasoning
82.77%Artificial Analysis
not published
Terminal-Bench Science 0.1Agentic science11.90%Artificial Analysisnot published
GDPval-AA v2.1Knowledge work1,663.25Artificial Analysisnot published
AA-Briefcase v1.1Knowledge work1,624.46Artificial Analysisnot published
Humanity's Last ExamMultidisciplinary reasoning43.10%Artificial Analysisnot published
Artificial Analysis Intelligence IndexComposite45Artificial 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.