GPT-5.2 vs Qwen4 Max
Qwen4 Max leads GPT-5.2 on coding, 74.0 to 56.0 on the Vibe Coding Index, and Qwen4 Max costs 3.3x less per output token ($4.275 vs $14.00 per million). Updated November 22, 2026; 4 benchmarks have a result for both.
GPT-5.2 or Qwen4 Max: which to pick
Pick Qwen4 Max for coding: it scores 74.0 on the Vibe Coding Index against 56.0 for GPT-5.2.
GPT-5.2 is neither cheaper nor larger in context, and leads no benchmark both models were measured on.
- Shared results
- 4
- GPT-5.2 only
- 110
- Qwen4 Max only
- 22
- Like-for-like categories
- 1 of 9
Only benchmarks that tested both models count toward a pick; a result only one model has is shown but not compared.
Which one for your work
Each card is decided by one number both models carry. Where only one has it, the card says so and makes no call.
Coding
Vibe Coding Index, six public coding benchmarks
Qwen4 Max
Qwen4 Max scores 74.0 against 56.0.
3 and 4 benchmarks behind the scores
Intelligence
AA Intelligence Index
No call
Only Qwen4 Max has a AA Intelligence Index result, 42.9.
Directional only
Reasoning
GPQA Diamond
No call
Only GPT-5.2 has a GPQA Diamond result, 91.7%.
Directional only
Math
AIME
No call
Only GPT-5.2 has a AIME result, 96.9%.
Directional only
Knowledge
MMLU Pro
No call
Only GPT-5.2 has a MMLU Pro result, 86.2%.
Directional only
Web design
Design Arena: websites
No call
Only GPT-5.2 has a Design Arena: websites result, 1182.
Directional only
General
Vals Index
No call
Only Qwen4 Max has a Vals Index result, 63.2%.
Directional only
Long documents
Listed context window
Qwen4 Max
Qwen4 Max takes 1M tokens against 400K.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
Qwen4 Max
GPT-5.2 $0.0088, Qwen4 Max $0.0036 per turn.
List prices
Cost per solved task
One SWE-bench Verified task, solved
No call
Only GPT-5.2 has the number, $1.91.
Directional only
At a glance
- Coding index
- GPT-5.2 56.0 from 3 benchmarks, Qwen4 Max 74.0 from 4 benchmarks.
- Widest category
- Coding, 31.8 points apart on average over 4 shared benchmarks.
- Widest single gap
- LMArena WebDev: GPT-5.2 1419 against Qwen4 Max 1751, 47.4 points apart on a 0 to 100 scale.
- Output price
- GPT-5.2 $14.00 against Qwen4 Max $4.275 per million output tokens, 3.3x apart.
- Context window
- GPT-5.2 takes 400K tokens, Qwen4 Max 1M.
Benchmark by benchmark
First the score: boards each model leads, per category, where both have a result. Then every board, a category to a fold; bars are each result on a 0 to 100 scale, so compare the two in a row, not one row with another.
| Category | GPT-5.2 | Bars | Qwen4 Max | Ties |
|---|---|---|---|---|
| Coding | 0 | 4 | ||
| All shared boards | 0 | 4 |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index04
| Benchmark | GPT-5.2 | Qwen4 Max | Leader |
|---|---|---|---|
| SWE-bench Verified | 72.8% | Not measured | Only GPT-5.2 measured |
| SWE-bench Multilingual | 66.7% | Not measured | Only GPT-5.2 measured |
| LiveBench coding | Not measured | 69.8% | Only Qwen4 Max measured |
| LMArena WebDev | 1419 | 1751 | Qwen4 Max |
| LiveCodeBench (Vals AI) | 85.4% | Not measured | Only GPT-5.2 measured |
| SWE-bench Verified (Vals AI) | 75.8% | Not measured | Only GPT-5.2 measured |
| Terminal-Bench 4 (Vals AI) | Not measured | 48.5% | Only Qwen4 Max measured |
| Vibe Code Bench (Vals AI) | 53.5% | 86.7% | Qwen4 Max |
| LiveCodeBench (Artificial Analysis run) | 88.9% | Not measured | Only GPT-5.2 measured |
| Terminal-Bench Hard (Artificial Analysis run) | 47.0% | Not measured | Only GPT-5.2 measured |
| Terminal-Bench 4.0 (Artificial Analysis run) | Not measured | 48.5% | Only Qwen4 Max measured |
| FrontierCode 2.0 | Not measured | 26.5% | Only Qwen4 Max measured |
| Terminal-Bench | 64.9% | Not measured | Only GPT-5.2 measured |
| GSO-Bench | 27.5% | Not measured | Only GPT-5.2 measured |
| WeirdML | 72.2% | Not measured | Only GPT-5.2 measured |
| MirrorCode | 32.2% | Not measured | Only GPT-5.2 measured |
| ALE-Bench | 1293.5 | 1831.0 | Qwen4 Max |
| LMArena Text: coding | 1442 | Not measured | Only GPT-5.2 measured |
| LMArena Text: software and IT | 1434 | Not measured | Only GPT-5.2 measured |
| LMArena WebDev: React | Not measured | 1785 | Only Qwen4 Max measured |
| LMArena WebDev: HTML | 1429 | 1752 | Qwen4 Max |
| LMArena WebDev: from an image | Not measured | 1675 | Only Qwen4 Max measured |
| GameBench 2 | Not measured | 82.4 | Only Qwen4 Max measured |
| Terminal-Bench 2.0 | 64.9% | Not measured | Only GPT-5.2 measured |
| Neither model has a result on the other 28 boards here | |||
IntelligenceDirectional onlyPublic boards, shown as published, kept out of the index2 boards, none shared
| Benchmark | GPT-5.2 | Qwen4 Max | Leader |
|---|---|---|---|
| AA Intelligence Index | Not measured | 42.9 | Only Qwen4 Max measured |
| Epoch Capabilities Index | 154.5 | Not measured | Only GPT-5.2 measured |
Agentic workDirectional onlyPublic boards, shown as published, kept out of the index19 boards, none shared
| Benchmark | GPT-5.2 | Qwen4 Max | Leader |
|---|---|---|---|
| Design Arena: agents building web apps | 1054 | Not measured | Only GPT-5.2 measured |
| Design Arena: agents building full-stack apps | 1020 | Not measured | Only GPT-5.2 measured |
| Design Arena: agents building mobile apps | 1086 | Not measured | Only GPT-5.2 measured |
| Design Arena: agents building Android apps | 1043 | Not measured | Only GPT-5.2 measured |
| Design Arena: agents building Godot games | 1142 | Not measured | Only GPT-5.2 measured |
| LiveBench research | Not measured | 84.7% | Only Qwen4 Max measured |
| LiveBench browser | Not measured | 80.7% | Only Qwen4 Max measured |
| LiveBench long agentic | Not measured | 46.6% | Only Qwen4 Max measured |
| Tax Agent Bench (Vals AI) | Not measured | 40.4% | Only Qwen4 Max measured |
| Legal Research Bench (Vals AI) | Not measured | 41.4% | Only Qwen4 Max measured |
| Harvey's Legal Agent Benchmark (Vals AI) | Not measured | 21.3% | Only Qwen4 Max measured |
| τ²-Bench Telecom | 84.8% | Not measured | Only GPT-5.2 measured |
| METR time horizons | 75.3% | Not measured | Only GPT-5.2 measured |
| Remote Labor Index | 2.5% | Not measured | Only GPT-5.2 measured |
| GDPval | 49.7% | Not measured | Only GPT-5.2 measured |
| DeepResearch Bench | 41.1% | Not measured | Only GPT-5.2 measured |
| Vending-Bench 2 | $3,591 | Not measured | Only GPT-5.2 measured |
| Berkeley Function Calling Leaderboard (BFCL) | 55.9% | Not measured | Only GPT-5.2 measured |
| τ³-Banking | 32.2% | Not measured | Only GPT-5.2 measured |
| Neither model has a result on the other 25 boards here | |||
ReasoningDirectional onlyPublic boards, shown as published, kept out of the index21 boards, none shared
| Benchmark | GPT-5.2 | Qwen4 Max | Leader |
|---|---|---|---|
| GPQA Diamond | 91.7% | Not measured | Only GPT-5.2 measured |
| LiveBench reasoning | Not measured | 92.2% | Only Qwen4 Max measured |
| MMMU Pro (Vals AI) | 86.7% | Not measured | Only GPT-5.2 measured |
| GPQA Diamond (Artificial Analysis run) | 90.3% | Not measured | Only GPT-5.2 measured |
| Humanity's Last Exam (Artificial Analysis run) | 37.7% | Not measured | Only GPT-5.2 measured |
| HLE-Diamond (Artificial Analysis run) | Not measured | 44.9% | Only Qwen4 Max measured |
| CritPt v2, hard subset (Artificial Analysis run) | Not measured | 43.0% | Only Qwen4 Max measured |
| GPQA Diamond (Epoch AI run) | 91.4% | Not measured | Only GPT-5.2 measured |
| ARC-AGI | 86.2% | Not measured | Only GPT-5.2 measured |
| ARC-AGI-2 | 52.9% | Not measured | Only GPT-5.2 measured |
| SimpleBench | 45.8% | Not measured | Only GPT-5.2 measured |
| Chess puzzles | 48.0% | Not measured | Only GPT-5.2 measured |
| Mystery game puzzles | 23.0% | Not measured | Only GPT-5.2 measured |
| EBR-bench | 23.0% | Not measured | Only GPT-5.2 measured |
| VPCT | 84.0% | Not measured | Only GPT-5.2 measured |
| LMArena Text: hard prompts | 1428 | Not measured | Only GPT-5.2 measured |
| LMArena Text: expert | 1444 | Not measured | Only GPT-5.2 measured |
| LMArena Text: hard prompts in English | 1431 | Not measured | Only GPT-5.2 measured |
| LMArena Vision: diagrams | 1265 | Not measured | Only GPT-5.2 measured |
| LMArena Vision: homework | 1276 | Not measured | Only GPT-5.2 measured |
| Kagi LLM Benchmark | 73.3% | Not measured | Only GPT-5.2 measured |
| Neither model has a result on the other 5 boards here | |||
MathDirectional onlyPublic boards, shown as published, kept out of the index10 boards, none shared
| Benchmark | GPT-5.2 | Qwen4 Max | Leader |
|---|---|---|---|
| AIME | 96.9% | Not measured | Only GPT-5.2 measured |
| MGSM (Vals AI) | 94.0% | Not measured | Only GPT-5.2 measured |
| AIME 2025 (Artificial Analysis run) | 99.0% | Not measured | Only GPT-5.2 measured |
| OTIS Mock AIME | 96.1% | Not measured | Only GPT-5.2 measured |
| FrontierMath, tiers 1 to 3 | 67.4% | Not measured | Only GPT-5.2 measured |
| FrontierMath, tier 4 | 31.7% | Not measured | Only GPT-5.2 measured |
| ProofBench | 15.0% | Not measured | Only GPT-5.2 measured |
| LMArena Text: math | 1440 | Not measured | Only GPT-5.2 measured |
| LMArena Text: mathematical work | 1442 | Not measured | Only GPT-5.2 measured |
| MathArena | 30.7% | Not measured | Only GPT-5.2 measured |
| Neither model has a result on the other 4 boards here | |||
KnowledgeDirectional onlyPublic boards, shown as published, kept out of the index15 boards, none shared
| Benchmark | GPT-5.2 | Qwen4 Max | Leader |
|---|---|---|---|
| MMLU Pro | 86.2% | Not measured | Only GPT-5.2 measured |
| LegalBench (Vals AI) | 82.8% | Not measured | Only GPT-5.2 measured |
| CaseLaw (Vals AI) | 66.0% | Not measured | Only GPT-5.2 measured |
| CorpFin (Vals AI) | 65.9% | Not measured | Only GPT-5.2 measured |
| TaxEval (Vals AI) | 75.8% | Not measured | Only GPT-5.2 measured |
| MortgageTax (Vals AI) | 67.1% | Not measured | Only GPT-5.2 measured |
| MedQA (Vals AI) | 94.1% | Not measured | Only GPT-5.2 measured |
| MedCode (Vals AI) | 49.8% | Not measured | Only GPT-5.2 measured |
| MMLU Pro (Artificial Analysis run) | 87.4% | Not measured | Only GPT-5.2 measured |
| AA-Omniscience | Not measured | 27.7 | Only Qwen4 Max measured |
| SimpleQA Verified | 37.1% | Not measured | Only GPT-5.2 measured |
| LMArena Text: science | 1425 | Not measured | Only GPT-5.2 measured |
| LMArena Text: medicine and healthcare | 1425 | Not measured | Only GPT-5.2 measured |
| LMArena Text: legal and government | 1428 | Not measured | Only GPT-5.2 measured |
| LMArena Text: business and finance | 1414 | Not measured | Only GPT-5.2 measured |
| Neither model has a result on the other 6 boards here | |||
Web designDirectional onlyPublic boards, shown as published, kept out of the index9 boards, none shared
| Benchmark | GPT-5.2 | Qwen4 Max | Leader |
|---|---|---|---|
| Design Arena: websites | 1182 | Not measured | Only GPT-5.2 measured |
| Design Arena: UI components | 1174 | Not measured | Only GPT-5.2 measured |
| Design Arena: game dev | 1167 | Not measured | Only GPT-5.2 measured |
| Design Arena: data viz | 1202 | Not measured | Only GPT-5.2 measured |
| Design Arena: 3D | 1046 | Not measured | Only GPT-5.2 measured |
| Design Arena: all code categories | 1151 | Not measured | Only GPT-5.2 measured |
| Design Arena: SVG | 1123 | Not measured | Only GPT-5.2 measured |
| Design Arena: ASCII art | 1138 | Not measured | Only GPT-5.2 measured |
| LiveBench design | Not measured | 79.1% | Only Qwen4 Max measured |
GeneralDirectional onlyPublic boards, shown as published, kept out of the index36 boards, none shared
| Benchmark | GPT-5.2 | Qwen4 Max | Leader |
|---|---|---|---|
| Vals Index | Not measured | 63.2% | Only Qwen4 Max measured |
| LiveBench data analysis | Not measured | 83.4% | Only Qwen4 Max measured |
| MedScribe (Vals AI) | 84.4% | Not measured | Only GPT-5.2 measured |
| IFBench | 75.4% | Not measured | Only GPT-5.2 measured |
| Long context reasoning (AA-LCR) | 82.7% | Not measured | Only GPT-5.2 measured |
| GDPval-AA | Not measured | 1809 | Only Qwen4 Max measured |
| DTBench | 90.9% | Not measured | Only GPT-5.2 measured |
| LMCA | 43.9% | Not measured | Only GPT-5.2 measured |
| CL-bench | 18.2% | Not measured | Only GPT-5.2 measured |
| Furniture assembly | 38.3% | Not measured | Only GPT-5.2 measured |
| LMArena Text | 1416 | Not measured | Only GPT-5.2 measured |
| LMArena Text: instruction following | 1409 | Not measured | Only GPT-5.2 measured |
| LMArena Text: creative writing | 1375 | Not measured | Only GPT-5.2 measured |
| LMArena Vision | 1246 | Not measured | Only GPT-5.2 measured |
| LMArena Text: English | 1422 | Not measured | Only GPT-5.2 measured |
| LMArena Text: non-English | 1405 | Not measured | Only GPT-5.2 measured |
| LMArena Text: Chinese | 1448 | Not measured | Only GPT-5.2 measured |
| LMArena Text: German | 1442 | Not measured | Only GPT-5.2 measured |
| LMArena Text: Japanese | 1419 | Not measured | Only GPT-5.2 measured |
| LMArena Text: Korean | 1375 | Not measured | Only GPT-5.2 measured |
| LMArena Text: French | 1454 | Not measured | Only GPT-5.2 measured |
| LMArena Text: Spanish | 1412 | Not measured | Only GPT-5.2 measured |
| LMArena Text: Russian | 1415 | Not measured | Only GPT-5.2 measured |
| LMArena Text: Polish | 1425 | Not measured | Only GPT-5.2 measured |
| LMArena Text: multi-turn | 1419 | Not measured | Only GPT-5.2 measured |
| LMArena Text: longer queries | 1412 | Not measured | Only GPT-5.2 measured |
| LMArena Text: writing and language | 1392 | Not measured | Only GPT-5.2 measured |
| LMArena Text: entertainment and media | 1378 | Not measured | Only GPT-5.2 measured |
| LMArena Vision: English | 1244 | Not measured | Only GPT-5.2 measured |
| LMArena Vision: Chinese | 1305 | Not measured | Only GPT-5.2 measured |
| LMArena Vision: reading text | 1257 | Not measured | Only GPT-5.2 measured |
| LMArena Vision: humor | 1227 | Not measured | Only GPT-5.2 measured |
| LMArena Vision: captioning | 1265 | Not measured | Only GPT-5.2 measured |
| LMArena Vision: recognizing things | 1232 | Not measured | Only GPT-5.2 measured |
| LMArena Vision: creative writing | 1226 | Not measured | Only GPT-5.2 measured |
| LMArena Document | 1404 | Not measured | Only GPT-5.2 measured |
| Neither model has a result on the other 13 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | GPT-5.2 | Qwen4 Max | Leader |
|---|---|---|---|
| Neither model has a result on 5 boards here | |||
What each workload costs
Three fixed token mixes at each model's listed price, so per-token rates become a bill.
Chat turn
1K input + 500 output tokens
Qwen4 Max costs less
Repository review
50K input + 3K output tokens
Qwen4 Max costs less
Agent loop
200K input + 10K output tokens
Qwen4 Max costs less
Listed OpenRouter rates, without cache or batch discounts. A workload that does not fit a model's context window would need splitting.
Specification differences
The leading value in a row is marked; a row with a missing value names no leader. Our two rows are marked in red.
| Specification | OpenAI GPT-5.2 | Qwen4 Max |
|---|---|---|
| Maker | OpenAI | Qwen |
| Weights | No public weights listed | No public weights listed |
| Released | December 10, 2025 | November 16, 2026 |
| Context window | 400K | 1M |
| Input $/Mtok | $1.75 | $1.425 |
| Output $/Mtok | $14.00 | $4.275 |
| Coding index | 56.0 | 74.0 |
| Independent measurements behind it | 3 | 4 |
| Cost per solved task | $1.91 | Not computable |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is GPT-5.2 better than Qwen4 Max for coding?
Pick Qwen4 Max for coding: it scores 74.0 on the Vibe Coding Index against 56.0 for GPT-5.2. GPT-5.2 is neither cheaper nor larger in context, and leads no benchmark both models were measured on.
Which is cheaper, GPT-5.2 or Qwen4 Max?
GPT-5.2 costs $1.75 per million input tokens and $14.00 per million output tokens; Qwen4 Max costs $1.425 and $4.275. Cost per solved task is known for GPT-5.2 only.
Which has the bigger context window, GPT-5.2 or Qwen4 Max?
Qwen4 Max, with 1M tokens against 400K.
How many benchmarks is the GPT-5.2 vs Qwen4 Max comparison based on?
4 benchmarks have a result for both models as of November 22, 2026. GPT-5.2's coding index rests on 3 independent benchmarks and Qwen4 Max's on 4. Where only one model has a result, the page says so instead of naming a winner.