AIEvals

Models

Qwen3.8 2.4T A95B

Specifications and API prices for Qwen3.8 2.4T A95B, recorded by Artificial Analysis.

AlibabaReleased 12 August 2026Context 262k tokensOutput Not published in the checked sourceOpen weights$2.00 in / $6.00 out per Mtok

Family Qwen3.816 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
Artificial Analysis Intelligence IndexArtificial Analysis4025of 42no list readClaude Opus 5.518 pts
Terminal-Bench 4.0Artificial Analysis11.11%34of 40no list readClaude Sonnet 5.552.53 pts
τ³-BankingArtificial Analysis49.07%4of 30no list readQwen3.8 Max2.27 pts
MLCR-AAArtificial Analysis0.00%37of 37no list readClaude Sonnet 5.575.00 pts
EnterpriseOps-Gym-AAArtificial Analysis47.36%7of 22no list readClaude Fable 53.76 pts
AutomationBench-AAArtificial Analysis57.25%22of 40no list readGemini 4 Argon20.26 pts
SciCodeArtificial Analysis54.05%28of 40no list readClaude Opus 5.512.85 pts
GDP.pdfArtificial Analysis15.00%31of 40no list readGPT-6.1 Sol16.00 pts
CritPtArtificial Analysis20.00%20of 42no list readGPT-5.6 Sol12.29 pts
AA-OmniscienceArtificial Analysis4.3231of 42no list readClaude Opus 5.542.1
AA-LCR v1.1Artificial Analysis80.33%26of 42no list readKimi K38.34 pts
GDPval-AA v2.1Artificial Analysis1,597.2717of 42no list readClaude Opus 5.5248.9
AA-Briefcase v1.1Artificial Analysis1,439.6722of 41no list readClaude Opus 5.5382.18
Humanity's Last ExamArtificial Analysis42.45%29of 42no list readClaude Opus 5.518.90 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%
Gemini 3.8 FlashGoogle4119.70%
Muse Spark 1.2Meta407.07%
Qwen3.8 2.4T A95BAlibaba4011.11%
Qwen3.8-Flash-NextAlibaba4025.25%
Qwen3.8 MaxAlibaba4018.69%
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 2.4T A95B$2.1639.9 tok/s

Results by benchmark

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

BenchmarkIndependent runAs reported by a publisher
τ³-BankingTool use49.07%Artificial Analysisnot published
MLCR-AAMedical reasoning0.00%Artificial Analysisnot published
EnterpriseOps-Gym-AAEnterprise workflows47.36%Artificial Analysisnot published
AutomationBench-AAWorkflow automation57.25%Artificial Analysisnot published
SciCodeScientific coding54.05%Artificial Analysisnot published
GDP.pdfDocument reasoning15.00%Artificial Analysisnot published
CritPtScientific reasoning20.00%Artificial Analysisnot published
AA-OmniscienceFactual knowledge4.32Artificial Analysisnot published
AA-LCR v1.1Long-context reasoning80.33%Artificial Analysisnot published
Terminal-Bench 4.0Agentic coding11.11%Artificial Analysisnot published
Terminal-Bench 2.1Agentic coding
82.02%Artificial Analysis
not published
GPQA DiamondScience reasoning
93.54%Artificial Analysis
not published
GDPval-AA v2.1Knowledge work1,597.27Artificial Analysisnot published
AA-Briefcase v1.1Knowledge work1,439.67Artificial Analysisnot published
Humanity's Last ExamMultidisciplinary reasoning42.45%Artificial Analysisnot published
Artificial Analysis Intelligence IndexComposite40Artificial 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.