Qwen3.8 Max vs Qwen4 Max
Full profiles: Qwen3.8 Max (Alibaba) and Qwen4 Max (Alibaba): every benchmark, price and rank.
Updated · Data from 13 sources
As of 19 Nov 2026, there are not enough shared benchmarks for a verdict, and Qwen4 Max is 1.3× cheaper at $2.25 vs $3.00 per 1M tokens blended. Choose Qwen3.8 Max for coding (61 vs 55); choose Qwen4 Max for reasoning (59 vs 51) and knowledge (59 vs 58).
Qwen4 Max was released on 16 Nov 2026 and has no BenchLeader Index yet (it needs five independent results across two categories). So this page gives no quality verdict: it compares what is published so far for both, price, context window, speed and the benchmarks both have results on. First results are tracked on Qwen4 Max's page.
Key facts (as of 19 Nov 2026)
- Qwen3.8 Max costs $3.00 per 1M tokens blended (list price, 19 Nov 2026).
- Qwen4 Max costs $2.25 per 1M tokens blended (list price, 19 Nov 2026).
- Of the 17 benchmarks both models have, Qwen4 Max leads 15 (BenchLeader, 19 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 case | Pick | Why | Confidence |
|---|---|---|---|
| Coding | Qwen3.8 Max | 61 vs 55 category score; FrontierCode 2.0 (AA) 23.4% vs 26.4% | Medium |
| Reasoning | Qwen4 Max | 59 vs 51 category score; HLE-Diamond (AA) 44.9% vs 24.0% | High |
| Knowledge | Qwen4 Max | 59 vs 58 category score; AA-Omniscience v1.1 27.7 vs 10.5 | Medium |
| Chat quality (human preference) | – | Only Qwen3.8 Max has an Arena rating (68 category score) | Not comparable |
| Lowest cost | Qwen4 Max | $2.25 vs $3.00 per 1M blended; agentic coding session $0.108 vs $0.144 | – |
| Fastest first answer | Qwen3.8 Max | 93.1 s (Qwen3.8 Max) vs 139.2 s (Qwen4 Max) to the first answer token | – |
| High volume on a budget | Neither flagship | A smaller model is usually the better fit: Cheapest good LLMs, Fastest LLMs | – |
Key differences
- Not enough shared benchmarks for a verdict.
- Qwen3.8 Max is stronger in coding.
- Qwen4 Max is stronger in knowledge, reasoning.
- Qwen4 Max is 1.3× cheaper ($2.25 vs $3.00 per 1M blended).
At a glance
| Spec | Qwen3.8 Max | Qwen4 Max |
|---|---|---|
| Developer | Alibaba | Alibaba |
| Released | 3 Aug 2026 | 16 Nov 2026 |
| Context window (tokens) | 1,000,000 | 991,808 |
| Compared at | Qwen3.8 Max | Qwen4 Max |
| Open weights | No | No |
| Price in / out / cached input per 1M | $2.00 / $6.00 / $0.250 | $1.50 / $4.50 / $0.071 |
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.
| Category | Qwen3.8 Max | Qwen4 Max | Leader |
|---|---|---|---|
| Reasoning | 50.9 | 58.5 | Qwen4 Max |
| Coding | 61.0 | 55.1 | Qwen3.8 Max |
| Agents & tools | 65.0 | – | not comparable |
| Maths | 64.9 | – | not comparable |
| Knowledge | 58.0 | 59.0 | Qwen4 Max |
| Human preference | 67.5 | – | not comparable |
| Multimodal | 65.5 | – | not comparable |
| Long context | 64.8 | – | not comparable |
| Composite | 43.3 | – | not 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.
Not drawn, measured for fewer than two of these models: Agents & tools, Maths, Instruction following, Long context, Multimodal, Human preference.
Where sources disagree
Across the 17 benchmarks both have, Qwen3.8 Max leads 2, Qwen4 Max leads 15. Every overall index that rates both puts Qwen4 Max ahead.
| Overall index | Qwen3.8 Max | Qwen4 Max | Leader |
|---|---|---|---|
| AA Intelligence Index v5.0.1 | 33.6 | 42.9 | Qwen4 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.
| Measure | Qwen3.8 Max | Qwen4 Max |
|---|---|---|
| Blended price per 1M tokens | $3.00 | $2.25 |
| Input / output per 1M | $2.00 / $6.00 | $1.50 / $4.50 |
| Cached input per 1M | $0.250 | $0.071 |
| Price per task: chat reply | $0.0026 | $0.0020 |
| Price per task: summarize a 30-page report | $0.028 | $0.021 |
| Price per task: code edit | $0.021 | $0.016 |
| Price per task: agentic coding session | $0.144 | $0.108 |
| Price per task: structured extraction | $0.0052 | $0.0039 |
| Agentic coding session, input mostly cached | $0.065 | $0.044 |
| Cost per benchmark task (AA Index) | $10.02 | $3.27 |
| Output tokens per task (AA Index) | 171k | 105k |
| Of which reasoning | 120k | 69k |
| Time per task (AA Index) | 48 min | 30 min |
| Output speed | 60 tok/s | 59 tok/s |
| Time to first answer | 93.1 s | 139.2 s |
| Time to a complete answer, last measured 13 Nov 2026 | 282.4 s | – |
Per token is not per task: an average task of Artificial Analysis' Intelligence Index cost $10.02 on Qwen3.8 Max and $3.27 on Qwen4 Max, and Qwen3.8 Max wrote 1.6× as many output tokens for the same work, which a per-token price hides.
Each model is compared at its best-ranked reasoning-effort setting. Qwen3.8 Max scores 59.0 and takes 93 s to its first answer token; Qwen3.8 Max (xhigh) scores 56.4 in 116 s. Qwen4 Max and takes 139 s to its first answer token. 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
143 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.
| Metric | Publisher | Qwen3.8 Max | Qwen4 Max |
|---|---|---|---|
| BenchLeader Index | BenchLeader | 59.0 | – |
| Agents & tools score | BenchLeader | 65.0 | – |
| Coding score | BenchLeader | 61.0 | 55.1 |
| Composite score | BenchLeader | 43.3 | – |
| Human preference score | BenchLeader | 67.5 | – |
| Knowledge score | BenchLeader | 58.0 | 59.0 |
| Long context score | BenchLeader | 64.8 | – |
| Maths score | BenchLeader | 64.9 | – |
| Multimodal score | BenchLeader | 65.5 | – |
| Reasoning score | BenchLeader | 50.9 | 58.5 |
| Blended price $/M | BenchLeader | $3.00 | $2.25 |
| Output speed | BenchLeader | 60 tok/s | 59 tok/s |
| Time to first answer | BenchLeader | 93.1 s | 139.2 s |
| Context window | BenchLeader | 1,000,000 tokens | 991,808 tokens |
| Reasoning | |||
| GPQA Diamond | Epoch AI Benchmarking Hub | 92.7%† | – |
| Arena Hard Prompts | Arena (LMArena) | 1507 | – |
| LiveBench 2.0 Reasoning | LiveBench | 88.2% | 92.2%† |
| LiveBench 1.0 Reasoning | LiveBench | 88.2% | – |
| GPQA Diamond (AA) | Artificial Analysis | 92.8% | – |
| Humanity's Last Exam (AA) | Artificial Analysis | 43.1% | – |
| HLE-Diamond (AA) | Artificial Analysis | 24.0% | 44.9% |
| GPQA Diamond (Vals) | Vals AI | 93.7% | – |
| CritPt | Artificial Analysis | 20.0% | – |
| CritPt v2 hard subset (AA) | Artificial Analysis | 29.5% | 43.0% |
| MysteryMechanism | Vals AI | 23.9% | – |
| Arena Expert | Arena (LMArena) | 1514 | – |
| Chess Puzzles | Epoch AI Benchmarking Hub | 46.0%† | – |
| Mystery Game Puzzles | Epoch AI Benchmarking Hub | 38.0%† | – |
| LMCA | Epoch AI Benchmarking Hub | 46.2%† | – |
| DTBench | Epoch AI Benchmarking Hub | 92.0%† | – |
| Coding | |||
| SciCode | SciCode | 53.2% | – |
| Arena Coding | Arena (LMArena) | 1525 | – |
| Arena WebDev | Arena (LMArena) | 1675 | 1750† |
| LiveBench 2.0 Coding | LiveBench | 60.9% | 69.8%† |
| LiveBench 1.0 Coding | LiveBench | 72.9% | – |
| LiveCodeBench | Vals AI | 87.8% | – |
| IOI | Vals AI | 68.9% | – |
| SWE-bench (Vals) | Vals AI | 85.6% | – |
| Code Migration | Vals AI | 24.0% | – |
Show all 143 benchmarks
| Metric | Publisher | Qwen3.8 Max | Qwen4 Max |
|---|---|---|---|
| Coding | |||
| ProgramBench | Vals AI | 0.0% | – |
| SWE-bench Multimodal | Vals AI | 42.5% | – |
| Vibe Code Bench 1-100 | Vals AI | 12.8% | – |
| Vibe Code Bench v1.1 | Vals AI | 64.7% | 86.7%† |
| SWE Atlas v1.1 | Scale AI SEAL | 64.4% | – |
| SWE Atlas v1.1: Codebase QnA | Scale AI SEAL | 81.5% | – |
| SWE Atlas v1.1: Test Writing | Scale AI SEAL | 58.9% | – |
| SWE Atlas v1.1: Refactoring | Scale AI SEAL | 52.9% | – |
| MirrorCode | Epoch AI Benchmarking Hub | 52.2%† | – |
| DeepSWE v1.1 | Epoch AI Benchmarking Hub | 57.5%† | – |
| ALE-Bench | Epoch AI Benchmarking Hub | 1679.1† | – |
| FrontierSWE | Epoch AI Benchmarking Hub | 17.8%† | – |
| Terminal-Bench 4.0 (AA) | Artificial Analysis | 38.9% | 48.5% |
| Terminal-Bench 2.1 (AA) | Artificial Analysis | 88.8% | – |
| Startrise Frontend | Startrise | 71.3% | – |
| GameBench 2 Overall | DuelLab GameBench 2 | 54.2 | – |
| GameBench 2 by reasoning setting | DuelLab GameBench 2 | 69.6† | – |
| Design Arena Website | Design Arena | 1272 | – |
| FrontierCode 1.1 (AA) | Artificial Analysis | 43.2% | – |
| FrontierCode 2.0 (AA) | Artificial Analysis | 23.4% | 26.4% |
| Agents & tools | |||
| Terminal-Bench | Terminal-Bench | 27.0%† | – |
| APEX-Agents | Mercor | 63.3% | – |
| Arena Agent | Arena (LMArena) | -0.9 | 7.3† |
| LiveBench Long Agentic | LiveBench | 26.6% | 46.6%† |
| LiveBench Browser | LiveBench | 49.4% | 80.7%† |
| LiveBench Agentic Coding | LiveBench | 64.7% | – |
| Terminal-Bench 2.1 (Vals) | Vals AI | 67.4% | – |
| τ³-bench Banking | τ²-bench | 55.1%† | – |
| GDPval-AA v2.1, normalised score | Artificial Analysis | 58.3% | – |
| GDPval-AA v2.1 | Artificial Analysis | 1674 | 1809 |
| τ³-Banking (AA) | Artificial Analysis | 51.3% | – |
| ITBench SRE (AA) | Artificial Analysis | 40.3% | – |
| Analyst Agent (AA) | Artificial Analysis | 45.0% | 46.3% |
| APEX-Agents (AA) | Artificial Analysis | 42.4% | 59.1% |
| CyberBench | Vals AI | 28.6% | – |
| Finance Agent v2 | Vals AI | 50.6% | – |
| Finance Agent v2.1 | Vals AI | – | 66.7%† |
| Harvey's Legal Agent Benchmark v1.1 | Vals AI | – | 21.3%† |
| Harvey's Legal Agent Benchmark v1 | Vals AI | 10.4% | – |
| Legal Research Bench | Vals AI | 47.6% | 41.4%† |
| Robo-bench | Vals AI | 5.2%† | – |
| SkillsBench | Vals AI | 42.0% | – |
| Tax Agent Bench | Vals AI | 32.1% | 40.4%† |
| Terminal-Bench 4.0 (Vals) | Vals AI | 34.3% | 48.5%† |
| Terminal-Bench Science | Vals AI | 12.9% | – |
| Arena Search | Arena (LMArena) | 1227 | – |
| Arena Agent Task Completion | Arena (LMArena) | 3.6 | 8.8† |
| AutomationBench | Artificial Analysis | 56.2% | 72.6% |
| GDP.pdf (AA) | Artificial Analysis | 22.8% | 26.2% |
| Harvey LAB-AA v1.1 | Harvey | – | 9.4% |
| EnterpriseOps-Gym | Artificial Analysis | 47.6% | 48.5% |
| AA-Briefcase v1.1 | Artificial Analysis | 1604 | 1686 |
| GDPx | GDPx by SemiAnalysis | 3.5%† | – |
| Design Arena Agentic | Design Arena | 1255 | 1356 |
| Terminal-Bench-Science 0.1 (AA) | Artificial Analysis | 11.9% | 23.3% |
| MKA (AA) | Artificial Analysis | 37.4% | – |
| FACTS Search | 85.0% | – | |
| Maths | |||
| FrontierMath Tiers 1–3 | Epoch AI Benchmarking Hub | 74.7%† | – |
| FrontierMath Tier 4 | Epoch AI Benchmarking Hub | 46.3%† | – |
| OTIS Mock AIME | Epoch AI Benchmarking Hub | 100.0%† | – |
| ProofBench | Vals AI | 58.0% | – |
| LiveBench Mathematics | LiveBench | 91.3% | – |
| Arena Maths | Arena (LMArena) | 1501 | – |
| Knowledge | |||
| SimpleQA Verified | Epoch AI Benchmarking Hub | 47.3%† | – |
| LiveBench 2.0 Data Analysis | LiveBench | 78.4% | 83.4%† |
| LiveBench Research | LiveBench | 72.8% | 84.7%† |
| LiveBench 1.0 Data Analysis | LiveBench | 78.4% | – |
| AA-Omniscience v1.1 | Artificial Analysis | 10.5 | 27.7 |
| MMLU-Pro | Vals AI | 88.6% | – |
| LegalBench | Vals AI | 83.6% | – |
| CorpFin | Vals AI | 65.8% | – |
| TaxEval | Vals AI | 75.5% | – |
| Excel Modeling Benchmark | Vals AI | 60.1% | – |
| Excel Modeling Benchmark v1.1 | Vals AI | 66.0% | 70.8%† |
| MedCode | Vals AI | 40.7% | – |
| MedScribe | Vals AI | 85.0% | – |
| Public Benefits Bench v1.1 | Vals AI | 67.1% | – |
| Access to Justice Benchmark | Vals AI | 76.4% | – |
| Financial Coaching Bench | Vals AI | 75.6% | – |
| Immigration Bench | Vals AI | 70.1% | – |
| GDP.pdf | Epoch AI Benchmarking Hub | 42.8%† | – |
| AA-Omniscience v1.1: accuracy | Artificial Analysis | 29.5% | 43.7% |
| AA-Omniscience v1.1: non-hallucination | Artificial Analysis | 72.9% | 71.6% |
| FACTS Parametric | 30.8% | – | |
| Instruction following | |||
| LiveBench Language | LiveBench | 79.7% | – |
| LiveBench Instruction Following | LiveBench | 74.1% | – |
| Arena Instruction Following | Arena (LMArena) | 1478 | – |
| Human preference | |||
| Arena Text | Arena (LMArena) | 1484 | – |
| Arena Creative Writing | Arena (LMArena) | 1473 | – |
| Arena Multi-turn | Arena (LMArena) | 1495 | – |
| Arena Longer Queries | Arena (LMArena) | 1498 | – |
| Multimodal | |||
| Arena Vision | Arena (LMArena) | 1312 | – |
| LiveBench Design | LiveBench | 76.6% | 79.1%† |
| MMMU-Pro | Artificial Analysis | 82.8% | – |
| MMMU-Pro (Vals) | Vals AI | 88.0% | – |
| MortgageTax | Vals AI | 64.0% | – |
| SAGE | Vals AI | 81.8% | – |
| Humanity's Sixth Sense | Scale AI SEAL | 33.0%† | – |
| Furniture Assembly | Epoch AI Benchmarking Hub | 20.0%† | – |
| VIALS-AA | Handshake AI | 31.7% | – |
| Code2Video-AA | Artificial Analysis | 1021 | – |
| AA-3D | Artificial Analysis | 954 | – |
| FACTS Multimodal | 51.5% | – | |
| Long context | |||
| AA-LCR v1.1 | Artificial Analysis | 80.3% | – |
| MLCR | Artificial Analysis | 26.1% | 19.4% |
| FACTS Grounding | 73.8% | – | |
| Composite | |||
| Epoch Capabilities Index | Epoch AI Benchmarking Hub | 157.6 | – |
| LiveBench 2.0 | LiveBench | 60.9% | 73.1%† |
| LiveBench 1.0 | LiveBench | 78.5% | – |
| AA Intelligence Index v5.0.1 | Artificial Analysis | 33.6 | 42.9 |
| AA Intelligence Index v4.3.2 | Artificial Analysis | 45.4 | – |
| AA Progress Index v1 | Artificial Analysis | 323.7 | – |
| Vals Index | Vals AI | – | 63.2† |
| Vals Finance Index | Vals AI | – | 62.0%† |
| Vals RSI Index | Vals AI | 21.8%† | – |
| Vals Multimodal Index | Vals AI | 65.4% | – |
| Vals Legal Index | Vals AI | – | 70.7%† |
| Safety & honesty | |||
| DistressBench | Scale AI SEAL | 77.6% | – |
Data as of 2026-11-19. Best configuration of each model; every score links to its source on the model pages.
DuelLab GameBench 2, release gb2-20261114-ca113f475f5e (https://benchmarks.duellab.org/releases/gb2-20261114-ca113f475f5e/), CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Scores mapped to BenchLeader model names; not endorsed by DuelLab.
What changed
- Qwen3.8 Max: BenchLeader Index 63.8 → 59.1.
- Qwen3.8 Max: 113 new results (FACTS Search, FACTS Grounding, FACTS Parametric…); 2 score revisions.
- Qwen3.8 Max: 1 score revision.
- Qwen3.8 Max: 1 score revision.
Related comparisons
- Qwen3.8 Max vs Gemini 4 Carbon
- Qwen4 Max vs Inkling 2 Small
- Qwen3.8 Max vs Qwen3.8-Flash-Next
- Qwen3.8 Max vs GPT-6 Astra
- Qwen3.8 Max vs GPT-6.1 Sol
- Qwen3.8 Max vs Gemini 4 Argon
- Qwen3.8 Max vs Claude Opus 5
- Qwen3.8 Max vs Muse Spark
Rankings: Best LLM overall · Best LLM for agents · Best cheap LLMs
Qwen3.8 Max vs Qwen4 Max: questions
- Qwen3.8 Max vs Qwen4 Max: which is cheaper, and will that change?
- Qwen4 Max is cheaper today: $2.25 against $3.00 per million tokens, blended at three input tokens per output token. No price change has been announced for either.
- Is Qwen3.8 Max or Qwen4 Max better for coding?
- Qwen3.8 Max scores higher in coding (61 vs 55 on the category index, where 50 is average). On FrontierCode 2.0 (AA), the top shared coding benchmark, it is 23.4% for Qwen3.8 Max against 26.4% for Qwen4 Max.
- Which is faster, Qwen3.8 Max or Qwen4 Max?
- Qwen3.8 Max streams faster (60 vs 59 output tokens per second, Artificial Analysis 7-day median).
- Which is better for long documents?
- Both take about 1M tokens: Qwen3.8 Max 1,000,000, Qwen4 Max 991,808.
- Which has lower latency?
- Qwen3.8 Max answers first: 93.1 s against 139.2 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.
More questions (2)
- Do independent benchmarks agree on Qwen3.8 Max vs Qwen4 Max?
- Across the 17 benchmarks both have, Qwen3.8 Max leads 2, Qwen4 Max leads 15. Every overall index that rates both puts Qwen4 Max ahead.
- When was Qwen4 Max released?
- Qwen4 Max was released on 16 Nov 2026.