GPT-5.2 vs Qwen3.5-35B-A3B
GPT-5.2 leads Qwen3.5-35B-A3B on the Artificial Analysis Intelligence Index, 30.4 to 19.3, and Qwen3.5-35B-A3B costs 10.8x less per output token ($1.30 vs $14.00 per million). Updated October 22, 2026; 44 benchmarks have a result for both.
GPT-5.2 or Qwen3.5-35B-A3B: which to pick
Only GPT-5.2 has a coding score, 55.9 on the Vibe Coding Index; Qwen3.5-35B-A3B has none, so the two cannot be ranked on coding.
Pick Qwen3.5-35B-A3B for output tokens 10.8x cheaper ($1.30 vs $14.00 per million) and work measured by τ²-Bench Telecom or LMArena Text: Chinese, where it leads.
- Shared results
- 44
- GPT-5.2 only
- 70
- Qwen3.5-35B-A3B only
- 1
- Like-for-like categories
- 7 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
No call
Only GPT-5.2 has a coding score, so no call is made.
One side not measured
Intelligence
AA Intelligence Index
GPT-5.2
GPT-5.2 leads, 30.4 against 19.3.
Same board, both measured
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, 1195.
Directional only
Long documents
Listed context window
GPT-5.2
GPT-5.2 takes 400K tokens against 262K.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
Qwen3.5-35B-A3B
GPT-5.2 $0.0088, Qwen3.5-35B-A3B $0.0008 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 55.9 from 3 benchmarks, Qwen3.5-35B-A3B not benchmarked.
- Widest category
- Coding, 18.1 points apart on average over 6 shared benchmarks.
- Widest single gap
- ALE-Bench: GPT-5.2 1293.5 against Qwen3.5-35B-A3B 221.8, 35.7 points apart on a 0 to 100 scale.
- Output price
- GPT-5.2 $14.00 against Qwen3.5-35B-A3B $1.30 per million output tokens, 10.8x apart.
- Context window
- GPT-5.2 takes 400K tokens, Qwen3.5-35B-A3B 262K.
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 | Qwen3.5-35B-A3B | Ties |
|---|---|---|---|---|
| Coding | 6 | 0 | ||
| Intelligence | 2 | 0 | ||
| Agentic work | 1 | 1 | ||
| Reasoning | 6 | 0 | ||
| Math | 3 | 0 | ||
| Knowledge | 4 | 0 | ||
| General | 20 | 1 | ||
| All shared boards | 42 | 2 |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index60
| Benchmark | GPT-5.2 | Qwen3.5-35B-A3B | 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 |
| LMArena WebDev | 1417 | 1254 | GPT-5.2 |
| 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 |
| Vibe Code Bench (Vals AI) | 53.5% | Not measured | Only GPT-5.2 measured |
| LiveCodeBench (Artificial Analysis run) | 89.4% | Not measured | Only GPT-5.2 measured |
| Terminal-Bench Hard (Artificial Analysis run) | 47.0% | 26.5% | GPT-5.2 |
| 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 |
| ALE-Bench | 1293.5 | 221.8 | GPT-5.2 |
| LMArena Text: coding | 1442 | 1410 | GPT-5.2 |
| LMArena Text: software and IT | 1434 | 1415 | GPT-5.2 |
| LMArena WebDev: React | Not measured | 1238 | Only Qwen3.5-35B-A3B measured |
| LMArena WebDev: HTML | 1427 | 1301 | GPT-5.2 |
| Neither model has a result on the other 23 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index20
| Benchmark | GPT-5.2 | Qwen3.5-35B-A3B | Leader |
|---|---|---|---|
| AA Intelligence Index | 30.4 | 19.3 | GPT-5.2 |
| Epoch Capabilities Index | 153.7 | 142.5 | GPT-5.2 |
Agentic workLike-for-likePublic boards, shown as published, kept out of the index11
| Benchmark | GPT-5.2 | Qwen3.5-35B-A3B | Leader |
|---|---|---|---|
| Design Arena: agents building web apps | 1086 | Not measured | Only GPT-5.2 measured |
| Design Arena: agents building full-stack apps | 1043 | Not measured | Only GPT-5.2 measured |
| Design Arena: agents building mobile apps | 1093 | Not measured | Only GPT-5.2 measured |
| Design Arena: agents building Android apps | 1059 | Not measured | Only GPT-5.2 measured |
| Design Arena: agents building Godot games | 1142 | Not measured | Only GPT-5.2 measured |
| τ²-Bench Telecom | 84.8% | 89.2% | Qwen3.5-35B-A3B |
| 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 | $463 | GPT-5.2 |
| Berkeley Function Calling Leaderboard (BFCL) | 55.9% | Not measured | Only GPT-5.2 measured |
| Neither model has a result on the other 29 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index60
| Benchmark | GPT-5.2 | Qwen3.5-35B-A3B | Leader |
|---|---|---|---|
| GPQA Diamond | 91.7% | Not measured | Only GPT-5.2 measured |
| MMMU Pro (Vals AI) | 86.7% | Not measured | Only GPT-5.2 measured |
| GPQA Diamond (Artificial Analysis run) | 90.3% | 84.5% | GPT-5.2 |
| Humanity's Last Exam (Artificial Analysis run) | 37.7% | 21.0% | GPT-5.2 |
| GPQA Diamond (Epoch AI run) | 91.4% | 83.5% | GPT-5.2 |
| 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 | 1400 | GPT-5.2 |
| LMArena Text: expert | 1445 | 1408 | GPT-5.2 |
| LMArena Text: hard prompts in English | 1431 | 1411 | GPT-5.2 |
| 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 4 boards here | |||
MathLike-for-likePublic boards, shown as published, kept out of the index30
| Benchmark | GPT-5.2 | Qwen3.5-35B-A3B | Leader |
|---|---|---|---|
| AIME | 96.9% | Not measured | Only GPT-5.2 measured |
| MGSM (Vals AI) | 94.0% | Not measured | Only GPT-5.2 measured |
| AA Math Index | 99.0 | 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% | 70.0% | GPT-5.2 |
| 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 | 1404 | GPT-5.2 |
| LMArena Text: mathematical work | 1442 | 1410 | GPT-5.2 |
| Neither model has a result on the other 3 boards here | |||
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index40
| Benchmark | GPT-5.2 | Qwen3.5-35B-A3B | 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 |
| SimpleQA Verified | 37.1% | Not measured | Only GPT-5.2 measured |
| LMArena Text: science | 1425 | 1413 | GPT-5.2 |
| LMArena Text: medicine and healthcare | 1425 | 1421 | GPT-5.2 |
| LMArena Text: legal and government | 1427 | 1399 | GPT-5.2 |
| LMArena Text: business and finance | 1414 | 1394 | GPT-5.2 |
| FACTS Benchmark Suite | 48.5% | Not measured | Only GPT-5.2 measured |
| Neither model has a result on the other 3 boards here | |||
Web designDirectional onlyPublic boards, shown as published, kept out of the index8 boards, none shared
| Benchmark | GPT-5.2 | Qwen3.5-35B-A3B | Leader |
|---|---|---|---|
| Design Arena: websites | 1195 | Not measured | Only GPT-5.2 measured |
| Design Arena: UI components | 1190 | Not measured | Only GPT-5.2 measured |
| Design Arena: game dev | 1195 | Not measured | Only GPT-5.2 measured |
| Design Arena: data viz | 1202 | Not measured | Only GPT-5.2 measured |
| Design Arena: 3D | 1079 | Not measured | Only GPT-5.2 measured |
| Design Arena: all code categories | 1170 | Not measured | Only GPT-5.2 measured |
| Design Arena: SVG | 1139 | Not measured | Only GPT-5.2 measured |
| Design Arena: ASCII art | 1152 | Not measured | Only GPT-5.2 measured |
| Neither model has a result on the other 1 board here | |||
GeneralLike-for-likePublic boards, shown as published, kept out of the index201
| Benchmark | GPT-5.2 | Qwen3.5-35B-A3B | Leader |
|---|---|---|---|
| SAGE (Vals AI) | 49.3% | Not measured | Only GPT-5.2 measured |
| MedScribe (Vals AI) | 84.4% | Not measured | Only GPT-5.2 measured |
| IFBench | 75.4% | 72.5% | GPT-5.2 |
| Long context reasoning (AA-LCR) | 82.7% | 72.0% | GPT-5.2 |
| DTBench | 90.9% | 80.0% | GPT-5.2 |
| LMCA | 43.9% | 29.5% | GPT-5.2 |
| 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 | 1395 | GPT-5.2 |
| LMArena Text: instruction following | 1409 | 1379 | GPT-5.2 |
| LMArena Text: creative writing | 1376 | 1347 | GPT-5.2 |
| LMArena Vision | 1246 | Not measured | Only GPT-5.2 measured |
| LMArena Text: English | 1422 | 1407 | GPT-5.2 |
| LMArena Text: non-English | 1405 | 1378 | GPT-5.2 |
| LMArena Text: Chinese | 1447 | 1457 | Qwen3.5-35B-A3B |
| LMArena Text: German | 1444 | 1367 | GPT-5.2 |
| LMArena Text: Japanese | 1420 | 1326 | GPT-5.2 |
| LMArena Text: Korean | 1374 | 1355 | GPT-5.2 |
| LMArena Text: French | 1454 | 1412 | GPT-5.2 |
| LMArena Text: Spanish | 1413 | 1390 | GPT-5.2 |
| LMArena Text: Russian | 1415 | 1376 | GPT-5.2 |
| LMArena Text: Polish | 1426 | 1385 | GPT-5.2 |
| LMArena Text: multi-turn | 1419 | 1390 | GPT-5.2 |
| LMArena Text: longer queries | 1413 | 1389 | GPT-5.2 |
| LMArena Text: writing and language | 1392 | 1365 | GPT-5.2 |
| LMArena Text: entertainment and media | 1378 | 1347 | GPT-5.2 |
| LMArena Vision: English | 1243 | 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 | 1267 | Not measured | Only GPT-5.2 measured |
| LMArena Vision: recognizing things | 1232 | Not measured | Only GPT-5.2 measured |
| LMArena Vision: creative writing | 1225 | Not measured | Only GPT-5.2 measured |
| LMArena Document | 1405 | Not measured | Only GPT-5.2 measured |
| Neither model has a result on the other 12 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | GPT-5.2 | Qwen3.5-35B-A3B | 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
Qwen3.5-35B-A3B costs less
Repository review
50K input + 3K output tokens
Qwen3.5-35B-A3B costs less
Agent loop
200K input + 10K output tokens
Qwen3.5-35B-A3B 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 | Qwen3.5-35B-A3B |
|---|---|---|
| Maker | OpenAI | Qwen |
| Weights | No public weights listed | Open weights |
| Released | December 10, 2025 | February 25, 2026 |
| Context window | 400K | 262K |
| Input $/Mtok | $1.75 | $0.1625 |
| Output $/Mtok | $14.00 | $1.30 |
| Coding index | 55.9 | Not benchmarked |
| Independent measurements behind it | 3 | 1 |
| Cost per solved task | $1.91 | Not computable |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is GPT-5.2 better than Qwen3.5-35B-A3B for coding?
Only GPT-5.2 has a coding score, 55.9 on the Vibe Coding Index; Qwen3.5-35B-A3B has none, so the two cannot be ranked on coding. Pick Qwen3.5-35B-A3B for output tokens 10.8x cheaper ($1.30 vs $14.00 per million) and work measured by τ²-Bench Telecom or LMArena Text: Chinese, where it leads.
Which is cheaper, GPT-5.2 or Qwen3.5-35B-A3B?
GPT-5.2 costs $1.75 per million input tokens and $14.00 per million output tokens; Qwen3.5-35B-A3B costs $0.1625 and $1.30. Cost per solved task is known for GPT-5.2 only.
Which has the bigger context window, GPT-5.2 or Qwen3.5-35B-A3B?
GPT-5.2, with 400K tokens against 262K.
How many benchmarks is the GPT-5.2 vs Qwen3.5-35B-A3B comparison based on?
44 benchmarks have a result for both models as of October 22, 2026. GPT-5.2's coding index rests on 3 independent benchmarks and Qwen3.5-35B-A3B's on 1. Where only one model has a result, the page says so instead of naming a winner.