BenchLeader

o3 vs Qwen3.8 Max

Full profiles: o3 (OpenAI) and Qwen3.8 Max (Alibaba): every benchmark, price and rank.

Updated · Data from 21 sources

As of 16 Nov 2026, o3 and Qwen3.8 Max are neck and neck on quality (60.4 vs 58.7 on the BenchLeader Index, inside the ±4.7 margin), and Qwen3.8 Max is 1.2× cheaper at $3.00 vs $3.50 per 1M tokens blended. Choose o3 for agents & tools (70 vs 59), reasoning (61 vs 52), and maths (71 vs 65) and fast first answers (5.2 s vs 119.6 s), where it also streams ~2× faster; choose Qwen3.8 Max for multimodal (66 vs 53), human preference (68 vs 61), and long context (65 vs 62).

Key facts (as of 16 Nov 2026)

  • o3 scores 60.4 on the BenchLeader Index and Qwen3.8 Max scores 58.7 (BenchLeader, 16 Nov 2026).
  • o3 costs $3.50 per 1M tokens blended (list price, 16 Nov 2026).
  • Qwen3.8 Max costs $3.00 per 1M tokens blended (list price, 16 Nov 2026).
  • o3 streams ~2× faster than Qwen3.8 Max (Artificial Analysis, 7-day median, 16 Nov 2026).
  • Of the 18 benchmarks both models have, Qwen3.8 Max leads 18 (BenchLeader, 16 Nov 2026).

Which to pick for…

The pick's numbers come first. Category scores on the BenchLeader scale (50 is the average model, 15 points one standard deviation). High confidence means a lead of half a standard deviation or more; under one point is called either.

Use casePickWhyConfidence
CodingQwen3.8 Max62 vs 60 category score; Arena Coding 1523 vs 1460Medium
Agents & tool useo370 vs 59 category score; Arena Search 1197 vs 1227High
Mathso371 vs 65 category score; Arena Maths 1448 vs 1505Medium
Reasoningo361 vs 52 category score; Arena Hard Prompts 1441 vs 1507High
KnowledgeQwen3.8 Max58 vs 56 category scoreMedium
Long documentsQwen3.8 Max65 vs 62 category score; AA-LCR v1.1 80.3% vs 74.7%Medium
Images and other mediaQwen3.8 Max66 vs 53 category score; Arena Vision 1312 vs 1214High
Chat quality (human preference)Qwen3.8 Max68 vs 61 on Arena votesMedium
Lowest costQwen3.8 Max$3.00 vs $3.50 per 1M blended; agentic coding session $0.144 vs $0.152–
Fastest first answero35.2 s (o3) vs 119.6 s (Qwen3.8 Max (max)) to the first answer token–
High volume on a budgetNeither flagshipA smaller model is usually the better fit: Cheapest good LLMs, Fastest LLMs–

Key differences

  • o3 and Qwen3.8 Max (max) are neck and neck on quality (60.4 vs 58.7, within the ±4.7 margin).
  • o3 is stronger in agents & tools, maths, reasoning.
  • Qwen3.8 Max (max) is stronger in coding, human preference, knowledge, long context, multimodal.
  • Qwen3.8 Max (max) is 1.2× cheaper ($3.00 vs $3.50 per 1M blended).
  • o3 streams 2.2× faster (131 vs 59 tokens per second).

At a glance

Speco3Qwen3.8 Max
DeveloperOpenAIAlibaba
Released16 Apr 20253 Aug 2026
Context window (tokens)200,0001,000,000
Compared ato3Qwen3.8 Max (max)
Open weightsNoNo
Price in / out / cached input per 1M$2.00 / $8.00 / $0.500$2.00 / $6.00 / $0.250

Category by category

Category scores on the BenchLeader scale: 50 is the average evaluated model, 15 points is one standard deviation. A gap under one point is called level.

Categoryo3Qwen3.8 Max (max)Leader
Reasoning60.951.5o3
Coding59.861.6Qwen3.8 Max
Agents & tools70.358.5o3
Maths70.665.2o3
Knowledge56.258.0Qwen3.8 Max
Instruction following65.2–not comparable
Human preference61.067.5Qwen3.8 Max
Multimodal52.865.7Qwen3.8 Max
Long context61.964.8Qwen3.8 Max
Composite–45.0not comparable

On one chart

Each domain on the index's own scale, where 50 is an average model. Click a domain to see the benchmarks behind it. Pick up to 5 models from the list.

2550 avg75ReasoningCodingAgents & toolsMathsKnowledgeLong contextMultimodalHumanpreferenceo3, Reasoning: 60.9o3, Coding: 59.8o3, Agents & tools: 70.3o3, Maths: 70.6o3, Knowledge: 56.2o3, Long context: 61.9o3, Multimodal: 52.8o3, Human preference: 61.0Qwen3.8 Max, Reasoning: 51.5Qwen3.8 Max, Coding: 61.6Qwen3.8 Max, Agents & tools: 58.5Qwen3.8 Max, Maths: 65.2Qwen3.8 Max, Knowledge: 58.0Qwen3.8 Max, Long context: 64.8Qwen3.8 Max, Multimodal: 65.7Qwen3.8 Max, Human preference: 67.5

Not drawn, measured for fewer than two of these models: Instruction following.

Model, click to add or removeIndex

Where sources disagree

Across the 18 benchmarks both have, o3 leads 0, Qwen3.8 Max leads 18. Every overall index that rates both puts Qwen3.8 Max ahead.

Overall indexo3Qwen3.8 Max (max)Leader
BenchLeader Index60.458.7neck and neck
Arena Text14321484Qwen3.8 Max

Each publisher runs its own tasks, harnesses and effort settings, so their overall verdicts can differ; BenchLeader reconciles them on one scale. † measured at another effort setting of the same model (the publisher did not test the one compared here).

Price, speed and latency

List prices blended at three input tokens per output token. Per-task prices use the token counts on the methodology page. Cost, tokens and time per benchmark task are measured: Artificial Analysis’ averages over its whole Intelligence Index run.

Measureo3Qwen3.8 Max (max)
Blended price per 1M tokens$3.50$3.00
Input / output per 1M$2.00 / $8.00$2.00 / $6.00
Cached input per 1M$0.500$0.250
Price per task: chat reply$0.0032$0.0026
Price per task: summarize a 30-page report$0.029$0.028
Price per task: code edit$0.024$0.021
Price per task: agentic coding session$0.152$0.144
Price per task: structured extraction$0.0056$0.0052
Agentic coding session, input mostly cached$0.085$0.065
Cost per benchmark task (AA Index)–$10.02
Output tokens per task (AA Index)–171k
Of which reasoning–120k
Time per task (AA Index)–48 min
Output speed131 tok/s59 tok/s
Time to first answer5.2 s119.6 s
Time to a complete answer, last measured 13 Nov 202617.3 s294.1 s

Each model is compared at its best-ranked reasoning-effort setting. o3 scores 60.4 and takes 5 s to its first answer token; o3 (high) scores 55.1 in 5 s, o3 (medium) scores 53.3 in 5 s. Qwen3.8 Max (max) scores 58.7 and takes 120 s to its first answer token; Qwen3.8 Max (xhigh) scores 56.6 in 3 s. Higher effort means more thinking before the answer: better scores, longer waits and more tokens billed. How reasoning effort changes scores and cost.

Every benchmark

189 benchmarks, grouped by category; each name links to the board and its leaders. † measured at another effort setting of the same model, shown for reference and not counted as a win.

MetricPublishero3Qwen3.8 Max (max)
BenchLeader IndexBenchLeader60.458.7
Agents & tools scoreBenchLeader70.358.5
Coding scoreBenchLeader59.861.6
Human preference scoreBenchLeader61.067.5
Instruction following scoreBenchLeader65.2–
Knowledge scoreBenchLeader56.258.0
Long context scoreBenchLeader61.964.8
Maths scoreBenchLeader70.665.2
Multimodal scoreBenchLeader52.865.7
Reasoning scoreBenchLeader60.951.5
Composite scoreBenchLeader–45.0
Blended price $/MBenchLeader$3.50$3.00
Output speedBenchLeader131 tok/s59 tok/s
Time to first answerBenchLeader5.2 s119.6 s
Context windowBenchLeader200,000 tokens1,000,000 tokens
Reasoning
GPQA DiamondEpoch AI Benchmarking Hub81.8%†92.7%†
Humanity's Last ExamScale AI / CAIS20.3%†–
SimpleBenchSimpleBench53.1%†–
Arena Hard PromptsArena (LMArena)14411507
LiveBench 2.0 ReasoningLiveBench–88.2%
LiveBench 1.0 ReasoningLiveBench–88.2%
GPQA Diamond (AA)Artificial Analysis82.7%92.8%
Humanity's Last Exam (AA)Artificial Analysis20.1%43.1%
HLE-Diamond (AA)Artificial Analysis–24.0%
GPQA Diamond (Vals)Vals AI84.1%†93.7%
Kagi LLM BenchmarkKagi LLM Benchmark67.6%–
GPQA Diamond (HELM)HELM Capabilities75.3%–
ARC-AGI-1ARC Prize60.8%†–
ARC-AGI-2ARC Prize6.5%†–
CritPtArtificial Analysis1.1%20.0%
CritPt v2 hard subset (AA)Artificial Analysis–29.5%
MysteryMechanismVals AI–23.9%
SciPredictScale AI SEAL17.9%–
Arena ExpertArena (LMArena)14481514
Chess PuzzlesEpoch AI Benchmarking Hub36.0%†46.0%†
Mystery Game PuzzlesEpoch AI Benchmarking Hub29.0%†38.0%†
CL-benchEpoch AI Benchmarking Hub17.8%†–
LMCAEpoch AI Benchmarking Hub39.7%†46.2%†
DTBenchEpoch AI Benchmarking Hub84.8%†92.0%†
ForecastBenchForecastBench64.3%–
Show all 189 benchmarks
MetricPublishero3Qwen3.8 Max (max)
Coding
SWE-bench Verified (Epoch)Epoch AI Benchmarking Hub62.3%†–
SciCodeSciCode–53.2%
WeirdMLWeirdML52.4%†–
GSO-BenchGSO-Bench8.8%†–
Arena CodingArena (LMArena)14601523
Arena WebDevArena (LMArena)–1676
LiveBench 2.0 CodingLiveBench–60.9%
LiveBench 1.0 CodingLiveBench–72.9%
LiveCodeBenchVals AI83.9%†87.8%
IOIVals AI–68.9%
SWE-bench (Vals)Vals AI–85.6%
Aider PolyglotAider polyglot leaderboard76.9%–
SWE-bench Verified (bash only)SWE-bench58.4%–
SWE-bench Verified (any scaffold)SWE-bench58.4%–
Code MigrationVals AI–24.0%
ProgramBenchVals AI–0.0%
SWE-bench MultimodalVals AI–42.5%
Vibe Code Bench 1-100Vals AI–12.8%
Vibe Code Bench v1.1Vals AI–64.7%
SWE Atlas v1.1Scale AI SEAL–64.4%
SWE Atlas v1.1: Codebase QnAScale AI SEAL–81.5%
SWE Atlas v1.1: Test WritingScale AI SEAL–58.9%
SWE Atlas v1.1: RefactoringScale AI SEAL–52.9%
MirrorCodeEpoch AI Benchmarking Hub–52.2%†
CadEvalEpoch AI Benchmarking Hub74.0%†–
DeepSWE v1.1Epoch AI Benchmarking Hub–57.5%†
ALE-BenchEpoch AI Benchmarking Hub933.5†1679.1†
FrontierSWEEpoch AI Benchmarking Hub–17.8%†
Terminal-Bench 4.0 (AA)Artificial Analysis–38.9%
Terminal-Bench 2.1 (AA)Artificial Analysis–88.8%
Startrise FrontendStartrise–71.3%
GameBench 2 OverallDuelLab GameBench 234.154.2†
GameBench 2 by reasoning settingDuelLab GameBench 230.4†69.6†
Design Arena WebsiteDesign Arena10281269
FrontierCode 1.1 (AA)Artificial Analysis–43.2%
FrontierCode 2.0 (AA)Artificial Analysis–23.4%
LiveCodeBench (AA)Artificial Analysis80.8%–
Agents & tools
Terminal-BenchTerminal-Bench–27.0%
GDPvalOpenAI30.8%†–
APEX-AgentsMercor–63.3%
Arena AgentArena (LMArena)–0.1
LiveBench Long AgenticLiveBench–26.6%
LiveBench BrowserLiveBench–49.4%
LiveBench Agentic CodingLiveBench–64.7%
Terminal-Bench HardArtificial Analysis37.1%–
τ²-Bench Telecom (AA)Artificial Analysis80.7%–
Terminal-Bench 2.1 (Vals)Vals AI–67.4%
BFCL OverallBerkeley Function Calling Leaderboard63.0%–
τ³-bench Bankingτ²-bench–55.1%†
GDPval-AA v2.1, normalised scoreArtificial Analysis–58.3%
GDPval-AA v2.1Artificial Analysis–1674
τ³-Banking (AA)Artificial Analysis–51.3%
ITBench SRE (AA)Artificial Analysis–40.3%
Analyst Agent (AA)Artificial Analysis–45.0%
APEX-Agents (AA)Artificial Analysis–42.4%
CyberBenchVals AI–28.6%
Finance Agent v2Vals AI–50.6%
Harvey's Legal Agent Benchmark v1Vals AI–10.4%
Legal Research BenchVals AI–47.6%
SkillsBenchVals AI–42.0%
Tax Agent BenchVals AI–32.1%
Terminal-Bench 4.0 (Vals)Vals AI–34.3%
Terminal-Bench ScienceVals AI–12.9%
Arena SearchArena (LMArena)11971227
Arena Agent Task CompletionArena (LMArena)–4.5
DeepResearch BenchDeepResearch Bench45.2%†–
METR Time HorizonsMETR63.6%–
AutomationBenchArtificial Analysis–56.2%
GDP.pdf (AA)Artificial Analysis–22.8%
EnterpriseOps-GymArtificial Analysis–47.6%
AA-Briefcase v1.1Artificial Analysis–1604
GDPxGDPx by SemiAnalysis–3.5%
Design Arena AgenticDesign Arena–1264
Terminal-Bench-Science 0.1 (AA)Artificial Analysis–11.9%
MKA (AA)Artificial Analysis–37.4%
BrowseComp-PlusBrowseComp-Plus50.5%–
FACTS SearchGoogle–85.0%
Maths
FrontierMath Tiers 1–3Epoch AI Benchmarking Hub33.3%†74.7%†
FrontierMath Tier 4Epoch AI Benchmarking Hub–46.3%†
OTIS Mock AIMEEpoch AI Benchmarking Hub83.9%†100.0%†
MATH Level 5Epoch AI Benchmarking Hub97.8%†–
ProofBenchVals AI–58.0%
LiveBench MathematicsLiveBench–91.3%
AIME (Vals)Vals AI85.3%†–
MGSMVals AI91.8%†–
IMO 2025MathArena16.7%†–
Omni-MATH (HELM)HELM Capabilities71.4%–
MATH 500Vals AI94.6%†–
Arena MathsArena (LMArena)14481505
AIME 2025 (AA)Artificial Analysis88.3%–
Knowledge
SimpleQA VerifiedEpoch AI Benchmarking Hub49.4%†47.3%†
LiveBench 2.0 Data AnalysisLiveBench–78.4%
LiveBench ResearchLiveBench–72.8%
LiveBench 1.0 Data AnalysisLiveBench–78.4%
AA-Omniscience v1.1Artificial Analysis–10.5
MMLU-ProVals AI85.6%†88.6%
LegalBenchVals AI83.8%†83.6%
CorpFinVals AI59.7%†65.8%
TaxEvalVals AI74.6%†75.5%
MedQAVals AI96.1%†–
PRBench FinanceScale AI SEAL47.7%–
PRBench LegalScale AI SEAL48.6%–
MultiNRCScale AI SEAL45.5%†–
MMLU-Pro (HELM)HELM Capabilities85.9%–
Excel Modeling BenchmarkVals AI–60.1%
Excel Modeling Benchmark v1.1Vals AI–66.0%
MedCodeVals AI47.3%†40.7%
MedScribeVals AI76.7%†85.0%
Public Benefits Bench v1.1Vals AI–67.1%
Access to Justice BenchmarkVals AI–76.4%
Financial Coaching BenchVals AI–75.6%
Immigration BenchVals AI–70.1%
GDP.pdfEpoch AI Benchmarking Hub–23.2%†
AA-Omniscience v1.1: accuracyArtificial Analysis–29.5%
AA-Omniscience v1.1: non-hallucinationArtificial Analysis–72.9%
FACTS ParametricGoogle–30.8%
Instruction following
LiveBench LanguageLiveBench–79.7%
LiveBench Instruction FollowingLiveBench–74.1%
IFBenchArtificial Analysis71.4%–
MultiChallengeScale AI SEAL56.6%†–
TutorBenchScale AI SEAL52.1%†–
IFEval (HELM)HELM Capabilities86.9%–
WildBench (HELM)HELM Capabilities86.1%–
Arena Instruction FollowingArena (LMArena)14031478
Human preference
Arena TextArena (LMArena)14321484
EQ-Bench Creative Writing v3EQ-Bench1675–
Arena Creative WritingArena (LMArena)13831472
Arena Multi-turnArena (LMArena)14201495
Arena Longer QueriesArena (LMArena)14111496
Lech Mazur Creative WritingLech Mazur83.9%†–
Multimodal
Arena VisionArena (LMArena)12141312
LiveBench DesignLiveBench–76.6%
MMMU-ProArtificial Analysis70.1%82.8%
VISTAScale AI SEAL50.1%†–
MMMU-Pro (Vals)Vals AI80.4%†88.0%
MortgageTaxVals AI65.7%†64.0%
SAGEVals AI–81.8%
Humanity's Sixth SenseScale AI SEAL–33.0%
VTBScale AI SEAL13.7%–
GeoBenchGeoBench60.0%†–
VPCTEpoch AI Benchmarking Hub52.0%†–
Furniture AssemblyEpoch AI Benchmarking Hub–20.0%†
VIALS-AAHandshake AI–31.7%
Code2Video-AAArtificial Analysis–1021
AA-3DArtificial Analysis–954
FACTS MultimodalGoogle–51.5%
Long context
Fiction.LiveBench 120kFiction.live100.0%†–
AA-LCR v1.1Artificial Analysis74.7%80.3%
MLCRArtificial Analysis–26.1%
FACTS GroundingGoogle–73.8%
Composite
Epoch Capabilities IndexEpoch AI Benchmarking Hub147.1157.6
LiveBench 2.0LiveBench–60.9%
LiveBench 1.0LiveBench–78.5%
AA Intelligence Index v5.0.1Artificial Analysis–33.6
AA Intelligence Index v4.3.2Artificial Analysis20.245.4
AA Progress Index v1Artificial Analysis258.3323.7
HELM Capabilities meanHELM Capabilities81.1%–
Vals RSI IndexVals AI–21.8%
Vals Multimodal IndexVals AI–65.4%
Safety & honesty
FORTRESSScale AI SEAL16.0%–
MASKScale AI SEAL84.5%†–
PropensityBenchScale AI SEAL10.5%–
DistressBenchScale AI SEAL–77.6%
Metadata
AA Openness IndexArtificial Analysis5.6–

Data as of 2026-11-16. Best configuration of each model; every score links to its source on the model pages.

DuelLab GameBench 2, release gb2-20261111-7b598bb13de4 (https://benchmarks.duellab.org/releases/gb2-20261111-7b598bb13de4/), CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Scores mapped to BenchLeader model names; not endorsed by DuelLab.

What changed

  • o3: BenchLeader Index 61.2 → 60.4.
  • o3: 1 score revision.
  • Qwen3.8 Max (max): 1 score revision.
  • Qwen3.8 Max (max): 1 score revision.
  • Qwen3.8 Max (max): 6 score revisions.
  • o3: 1 new result (Arena Search).
  • Qwen3.8 Max (max): 1 score revision.
  • Qwen3.8 Max (max): 1 new result (FACTS Search); 1 score revision.
  • o3: 1 score revision.
  • Qwen3.8 Max (max): 1 score revision; cached input price moved from $0.22/M to $0.25/M (13% up).

Rankings: Best LLM overall · Best LLM for agents · Best cheap LLMs

o3 vs Qwen3.8 Max: questions

Is o3 better than Qwen3.8 Max?
o3 and Qwen3.8 Max (max) are neck and neck on quality (60.4 vs 58.7, within the ±4.7 margin). The BenchLeader Index combines every independent quality benchmark; the two are within the margin of error as of 16 Nov 2026, so check the category scores for your use.
o3 vs Qwen3.8 Max: which is cheaper, and will that change?
Qwen3.8 Max is cheaper today: $3.00 against $3.50 per million tokens, blended at three input tokens per output token. No price change has been announced for either.
Is o3 or Qwen3.8 Max better for coding?
Qwen3.8 Max scores higher in coding (62 vs 60 on the category index, where 50 is average). On Arena Coding, the top shared coding benchmark, it is 1460 for o3 against 1523 for Qwen3.8 Max.
Which is faster, o3 or Qwen3.8 Max?
o3 streams ~2× faster (131 vs 59 output tokens per second, Artificial Analysis 7-day median).
Is o3 or Qwen3.8 Max better for agents and tool use?
o3 scores higher in agentic tasks (70 vs 59 on the category index, where 50 is average).
More questions (4)
Which is better for long documents?
Qwen3.8 Max scores higher on long-context benchmarks (65 vs 62). Qwen3.8 Max accepts more context: 1,000,000 against 200,000 tokens for o3.
Which has lower latency?
o3 answers first: 5.2 s against 119.6 s to the first answer token, as Artificial Analysis measures it. Reasoning effort drives most of this: a higher setting thinks longer before it answers.
Do independent benchmarks agree on o3 vs Qwen3.8 Max?
Across the 18 benchmarks both have, o3 leads 0, Qwen3.8 Max leads 18. Every overall index that rates both puts Qwen3.8 Max ahead.
When was Qwen3.8 Max released?
Qwen3.8 Max was released on 3 Aug 2026.