GPT-5 vs Qwen3.5 397B A17B
Updated October 21, 2026. 62 benchmarks have a result for both. Coding is ranked on the Vibe Coding Index; every other board is shown as published.
Decision reading
Pick GPT-5 for coding: it scores 72.2 on the Vibe Coding Index against 28.6 for Qwen3.5 397B A17B.
Pick Qwen3.5 397B A17B when output tokens cost 4.3x less, or when what matters is Terminal-Bench Hard (Artificial Analysis run) or LMArena Text: coding, where it leads.
- Shared results
- 62
- GPT-5 only
- 47
- Qwen3.5 397B A17B only
- 3
- Like-for-like categories
- 8 of 9
Shared results can support a head-to-head reading. Results only one model has describe coverage, not superiority.
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, four public coding benchmarks
GPT-5
GPT-5 scores 72.2 against 28.6.
3 and 1 benchmarks behind the scores
Intelligence
AA Intelligence Index
GPT-5
GPT-5 leads, 23.0 against 21.4.
Same board, both measured
Reasoning
GPQA Diamond
No call
Only GPT-5 has a GPQA Diamond result, 85.6%.
Directional only
Math
AIME
No call
Only GPT-5 has a AIME result, 93.4%.
Directional only
Knowledge
MMLU Pro
No call
Only GPT-5 has a MMLU Pro result, 86.5%.
Directional only
Web design
Design Arena: websites
Qwen3.5 397B A17B
Qwen3.5 397B A17B leads, 1191 against 1185.
Same board, both measured
Long documents
Listed context window
GPT-5
GPT-5 takes 400K tokens against 262K.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
Qwen3.5 397B A17B
GPT-5 $0.0063, Qwen3.5 397B A17B $0.0016 per turn.
List prices
Cost per solved task
One SWE-bench Verified task, solved
No call
Only GPT-5 has the number, $0.39.
Directional only
At a glance
- Coding index
- GPT-5 72.2 from 3 benchmarks, Qwen3.5 397B A17B 28.6 from 1 benchmark.
- Widest category
- Math, 10.7 points apart on average over 4 shared benchmarks.
- Widest single gap
- FrontierMath, tiers 1 to 3: GPT-5 55.4% against Qwen3.5 397B A17B 31.2%, 24.2 points apart on a 0 to 100 scale.
- Output price
- GPT-5 $10.00 against Qwen3.5 397B A17B $2.34 per million output tokens, 4.3x apart.
- Context window
- GPT-5 takes 400K tokens, Qwen3.5 397B A17B 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 | Bars | Qwen3.5 397B A17B | Ties |
|---|---|---|---|---|
| Coding | 3 | 3 | ||
| Intelligence | 2 | 0 | ||
| Agentic work | 0 | 1 | ||
| Reasoning | 1 | 8 | ||
| Math | 2 | 2 | ||
| Knowledge | 0 | 4 | ||
| Web design | 4 | 3 | ||
| General | 3 | 26 | ||
| All shared boards | 15 | 47 |
CodingLike-for-likeFour of these boards feed the Vibe Coding Index33
| Benchmark | GPT-5 | Qwen3.5 397B A17B | Leader |
|---|---|---|---|
| Aider polyglot | 88.0% | Not measured | Only GPT-5 measured |
| SWE-bench Verified | 74.4% | Not measured | Only GPT-5 measured |
| LMArena WebDev | 1418 | 1400 | GPT-5 |
| AA Coding Index | 37.8 | Not measured | Only GPT-5 measured |
| LiveCodeBench (Vals AI) | 85.9% | Not measured | Only GPT-5 measured |
| SWE-bench Verified (Vals AI) | 69.0% | Not measured | Only GPT-5 measured |
| Vibe Code Bench (Vals AI) | 20.1% | Not measured | Only GPT-5 measured |
| LiveCodeBench (Artificial Analysis run) | 84.6% | Not measured | Only GPT-5 measured |
| SciCode | Not measured | 44.8% | Only Qwen3.5 397B A17B measured |
| Terminal-Bench Hard (Artificial Analysis run) | 37.9% | 40.9% | Qwen3.5 397B A17B |
| Terminal-Bench | 49.6% | Not measured | Only GPT-5 measured |
| GSO-Bench | 6.9% | Not measured | Only GPT-5 measured |
| WeirdML | 60.7% | Not measured | Only GPT-5 measured |
| ALE-Bench | 807.6 | 621.9 | GPT-5 |
| LMArena Text: coding | 1436 | 1464 | Qwen3.5 397B A17B |
| LMArena Text: software and IT | 1420 | 1463 | Qwen3.5 397B A17B |
| LMArena WebDev: React | Not measured | 1388 | Only Qwen3.5 397B A17B measured |
| LMArena WebDev: HTML | 1427 | 1410 | GPT-5 |
| Neither model has a result on the other 20 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index20
| Benchmark | GPT-5 | Qwen3.5 397B A17B | Leader |
|---|---|---|---|
| AA Intelligence Index | 23.0 | 21.4 | GPT-5 |
| Epoch Capabilities Index | 150.0 | 146.8 | GPT-5 |
Agentic workLike-for-likePublic boards, shown as published, kept out of the index01
| Benchmark | GPT-5 | Qwen3.5 397B A17B | Leader |
|---|---|---|---|
| τ²-Bench Telecom | 86.5% | 95.6% | Qwen3.5 397B A17B |
| APEX-Agents | Not measured | 24.9% | Only Qwen3.5 397B A17B measured |
| METR time horizons | 69.6% | Not measured | Only GPT-5 measured |
| Remote Labor Index | 1.7% | Not measured | Only GPT-5 measured |
| GDPval | 34.8% | Not measured | Only GPT-5 measured |
| BALROG | 32.8% | Not measured | Only GPT-5 measured |
| DeepResearch Bench | 49.6% | Not measured | Only GPT-5 measured |
| Neither model has a result on the other 32 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index18
| Benchmark | GPT-5 | Qwen3.5 397B A17B | Leader |
|---|---|---|---|
| GPQA Diamond | 85.6% | Not measured | Only GPT-5 measured |
| MMMU Pro (Vals AI) | 81.5% | Not measured | Only GPT-5 measured |
| GPQA Diamond (Artificial Analysis run) | 85.4% | 89.3% | Qwen3.5 397B A17B |
| Humanity's Last Exam (Artificial Analysis run) | 28.5% | 29.0% | Qwen3.5 397B A17B |
| GPQA Diamond (Epoch AI run) | 86.2% | 86.4% | Qwen3.5 397B A17B |
| ARC-AGI | 65.7% | Not measured | Only GPT-5 measured |
| ARC-AGI-2 | 9.9% | Not measured | Only GPT-5 measured |
| SimpleBench | 56.7% | Not measured | Only GPT-5 measured |
| Chess puzzles | 36.0% | Not measured | Only GPT-5 measured |
| Mystery game puzzles | 23.0% | 18.0% | GPT-5 |
| EBR-bench | 12.7% | Not measured | Only GPT-5 measured |
| VPCT | 66.0% | Not measured | Only GPT-5 measured |
| LMArena Text: hard prompts | 1415 | 1448 | Qwen3.5 397B A17B |
| LMArena Text: expert | 1419 | 1462 | Qwen3.5 397B A17B |
| LMArena Text: hard prompts in English | 1419 | 1450 | Qwen3.5 397B A17B |
| LMArena Vision: diagrams | 1252 | 1277 | Qwen3.5 397B A17B |
| LMArena Vision: homework | 1278 | 1292 | Qwen3.5 397B A17B |
| Neither model has a result on the other 4 boards here | |||
MathLike-for-likePublic boards, shown as published, kept out of the index22
| Benchmark | GPT-5 | Qwen3.5 397B A17B | Leader |
|---|---|---|---|
| AIME | 93.4% | Not measured | Only GPT-5 measured |
| MATH 500 (Vals AI) | 96.0% | Not measured | Only GPT-5 measured |
| MGSM (Vals AI) | 92.8% | Not measured | Only GPT-5 measured |
| AA Math Index | 94.3 | Not measured | Only GPT-5 measured |
| AIME 2025 (Artificial Analysis run) | 94.3% | Not measured | Only GPT-5 measured |
| OTIS Mock AIME | 91.4% | 88.9% | GPT-5 |
| FrontierMath, tiers 1 to 3 | 55.4% | 31.2% | GPT-5 |
| FrontierMath, tier 4 | 22.0% | Not measured | Only GPT-5 measured |
| ProofBench | 18.0% | Not measured | Only GPT-5 measured |
| MATH level 5 | 98.1% | Not measured | Only GPT-5 measured |
| LMArena Text: math | 1407 | 1453 | Qwen3.5 397B A17B |
| LMArena Text: mathematical work | 1405 | 1454 | Qwen3.5 397B A17B |
| Neither model has a result on the other 1 board here | |||
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index04
| Benchmark | GPT-5 | Qwen3.5 397B A17B | Leader |
|---|---|---|---|
| MMLU Pro | 86.5% | Not measured | Only GPT-5 measured |
| LegalBench (Vals AI) | 86.0% | Not measured | Only GPT-5 measured |
| CaseLaw (Vals AI) | 66.5% | Not measured | Only GPT-5 measured |
| CorpFin (Vals AI) | 61.1% | Not measured | Only GPT-5 measured |
| TaxEval (Vals AI) | 73.4% | Not measured | Only GPT-5 measured |
| MortgageTax (Vals AI) | 65.5% | Not measured | Only GPT-5 measured |
| MedQA (Vals AI) | 96.3% | Not measured | Only GPT-5 measured |
| MedCode (Vals AI) | 49.6% | Not measured | Only GPT-5 measured |
| MMLU Pro (Artificial Analysis run) | 87.1% | Not measured | Only GPT-5 measured |
| SimpleQA Verified | 50.1% | Not measured | Only GPT-5 measured |
| GeoBench | 81.0% | Not measured | Only GPT-5 measured |
| LMArena Text: science | 1416 | 1457 | Qwen3.5 397B A17B |
| LMArena Text: medicine and healthcare | 1413 | 1455 | Qwen3.5 397B A17B |
| LMArena Text: legal and government | 1431 | 1441 | Qwen3.5 397B A17B |
| LMArena Text: business and finance | 1400 | 1435 | Qwen3.5 397B A17B |
| Neither model has a result on the other 2 boards here | |||
Web designLike-for-likePublic boards, shown as published, kept out of the index43
| Benchmark | GPT-5 | Qwen3.5 397B A17B | Leader |
|---|---|---|---|
| Design Arena: websites | 1185 | 1191 | Qwen3.5 397B A17B |
| Design Arena: UI components | 1179 | 1163 | GPT-5 |
| Design Arena: game dev | 1185 | 1141 | GPT-5 |
| Design Arena: data viz | 1221 | 1175 | GPT-5 |
| Design Arena: 3D | 1056 | 1161 | Qwen3.5 397B A17B |
| Design Arena: all code categories | 1168 | 1180 | Qwen3.5 397B A17B |
| Design Arena: SVG | 1188 | 1139 | GPT-5 |
| Design Arena: ASCII art | 1140 | Not measured | Only GPT-5 measured |
| Neither model has a result on the other 1 board here | |||
GeneralLike-for-likePublic boards, shown as published, kept out of the index326
| Benchmark | GPT-5 | Qwen3.5 397B A17B | Leader |
|---|---|---|---|
| SAGE (Vals AI) | 43.7% | Not measured | Only GPT-5 measured |
| MedScribe (Vals AI) | 83.7% | Not measured | Only GPT-5 measured |
| IFBench | 73.1% | 78.8% | Qwen3.5 397B A17B |
| Long context reasoning (AA-LCR) | 78.2% | 77.3% | GPT-5 |
| DTBench | 90.7% | 87.5% | GPT-5 |
| LMCA | 40.0% | 37.9% | GPT-5 |
| Fiction.LiveBench | 97.2% | Not measured | Only GPT-5 measured |
| Creative writing (Lech Mazur) | 86.0% | Not measured | Only GPT-5 measured |
| LMArena Text | 1406 | 1437 | Qwen3.5 397B A17B |
| LMArena Text: instruction following | 1388 | 1424 | Qwen3.5 397B A17B |
| LMArena Text: creative writing | 1365 | 1402 | Qwen3.5 397B A17B |
| LMArena Vision | 1232 | 1262 | Qwen3.5 397B A17B |
| LMArena Text: English | 1409 | 1443 | Qwen3.5 397B A17B |
| LMArena Text: non-English | 1397 | 1430 | Qwen3.5 397B A17B |
| LMArena Text: Chinese | 1422 | 1498 | Qwen3.5 397B A17B |
| LMArena Text: German | 1416 | 1445 | Qwen3.5 397B A17B |
| LMArena Text: Japanese | 1409 | 1429 | Qwen3.5 397B A17B |
| LMArena Text: Korean | 1360 | 1384 | Qwen3.5 397B A17B |
| LMArena Text: French | 1409 | 1462 | Qwen3.5 397B A17B |
| LMArena Text: Spanish | 1399 | 1441 | Qwen3.5 397B A17B |
| LMArena Text: Russian | 1407 | 1430 | Qwen3.5 397B A17B |
| LMArena Text: Polish | 1431 | 1447 | Qwen3.5 397B A17B |
| LMArena Text: multi-turn | 1426 | 1446 | Qwen3.5 397B A17B |
| LMArena Text: longer queries | 1399 | 1441 | Qwen3.5 397B A17B |
| LMArena Text: writing and language | 1384 | 1415 | Qwen3.5 397B A17B |
| LMArena Text: entertainment and media | 1379 | 1399 | Qwen3.5 397B A17B |
| LMArena Vision: English | 1231 | 1268 | Qwen3.5 397B A17B |
| LMArena Vision: Chinese | 1267 | 1314 | Qwen3.5 397B A17B |
| LMArena Vision: reading text | 1246 | 1271 | Qwen3.5 397B A17B |
| LMArena Vision: humor | 1219 | 1262 | Qwen3.5 397B A17B |
| LMArena Vision: captioning | 1212 | 1230 | Qwen3.5 397B A17B |
| LMArena Vision: recognizing things | 1263 | 1265 | Qwen3.5 397B A17B |
| LMArena Vision: creative writing | 1210 | 1260 | Qwen3.5 397B A17B |
| 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 | Qwen3.5 397B A17B | 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 397B A17B costs less
Repository review
50K input + 3K output tokens
Qwen3.5 397B A17B costs less
Agent loop
200K input + 10K output tokens
Qwen3.5 397B A17B 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 | Qwen3.5 397B A17B |
|---|---|---|
| Maker | OpenAI | Qwen |
| Weights | No public weights listed | Open weights |
| Released | August 7, 2025 | February 16, 2026 |
| Context window | 400K | 262K |
| Input $/Mtok | $1.25 | $0.39 |
| Output $/Mtok | $10.00 | $2.34 |
| Coding index | 72.2 | 28.6 |
| Independent measurements behind it | 3 | 1 |
| Cost per solved task | $0.39 | Not computable |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is GPT-5 better than Qwen3.5 397B A17B for coding?
Pick GPT-5 for coding: it scores 72.2 on the Vibe Coding Index against 28.6 for Qwen3.5 397B A17B. Pick Qwen3.5 397B A17B when output tokens cost 4.3x less, or when what matters is Terminal-Bench Hard (Artificial Analysis run) or LMArena Text: coding, where it leads.
Which is cheaper, GPT-5 or Qwen3.5 397B A17B?
GPT-5 costs $1.25 per million input tokens and $10.00 per million output tokens; Qwen3.5 397B A17B costs $0.39 and $2.34. Cost per solved task is known for GPT-5 only.
Which has the bigger context window, GPT-5 or Qwen3.5 397B A17B?
GPT-5, with 400K tokens against 262K.
How many benchmarks is the GPT-5 vs Qwen3.5 397B A17B comparison based on?
62 benchmarks have a result for both models as of October 21, 2026. GPT-5's coding index rests on 3 independent benchmarks and Qwen3.5 397B A17B's on 1. Where only one model has a result, the page says so instead of naming a winner.