Qwen3.5 397B A17B vs Qwen3 VL 235B A22B Instruct
Qwen3.5 397B A17B leads Qwen3 VL 235B A22B Instruct on the Artificial Analysis Intelligence Index, 21.4 to 9.9, and Qwen3 VL 235B A22B Instruct costs 1.2x less per output token ($1.90 vs $2.34 per million). Updated October 23, 2026; 45 benchmarks have a result for both.
Qwen3.5 397B A17B or Qwen3 VL 235B A22B Instruct: which to pick
Neither model has a coding score yet, so the two cannot be ranked on coding.
Pick Qwen3.5 397B A17B for work measured by LMArena Text: Korean or LMArena Vision: captioning, where it leads.
- Shared results
- 45
- Qwen3.5 397B A17B only
- 21
- Qwen3 VL 235B A22B Instruct only
- 4
- 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
Neither has a coding score yet.
One side not measured
Intelligence
AA Intelligence Index
Qwen3.5 397B A17B
Qwen3.5 397B A17B leads, 21.4 against 9.9.
Same board, both measured
Web design
Design Arena: websites
No call
Only Qwen3.5 397B A17B has a Design Arena: websites result, 1190.
Directional only
Long documents
Listed context window
No call
Both take 262K tokens.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
Qwen3 VL 235B A22B Instruct
Qwen3.5 397B A17B $0.0016, Qwen3 VL 235B A22B Instruct $0.0012 per turn.
List prices
At a glance
- Coding index
- Qwen3.5 397B A17B not benchmarked, Qwen3 VL 235B A22B Instruct not benchmarked.
- Widest category
- Agentic work, 41.5 points apart on average over 1 shared benchmark.
- Widest single gap
- τ²-Bench Telecom: Qwen3.5 397B A17B 95.6% against Qwen3 VL 235B A22B Instruct 54.1%, 41.5 points apart on a 0 to 100 scale.
- Output price
- Qwen3.5 397B A17B $2.34 against Qwen3 VL 235B A22B Instruct $1.90 per million output tokens, 1.2x apart.
- Context window
- Both take 262K tokens.
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 | Qwen3.5 397B A17B | Bars | Qwen3 VL 235B A22B Instruct | Ties |
|---|---|---|---|---|
| Coding | 3 | 0 | ||
| Intelligence | 1 | 0 | ||
| Agentic work | 1 | 0 | ||
| Reasoning | 7 | 0 | ||
| Math | 2 | 0 | ||
| Knowledge | 4 | 0 | ||
| General | 25 | 2 | ||
| All shared boards | 43 | 2 |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index30
| Benchmark | Qwen3.5 397B A17B | Qwen3 VL 235B A22B Instruct | Leader |
|---|---|---|---|
| LMArena WebDev | 1400 | Not measured | Only Qwen3.5 397B A17B measured |
| LiveCodeBench (Artificial Analysis run) | Not measured | 64.6% | Only Qwen3 VL 235B A22B Instruct measured |
| SciCode | 44.8% | Not measured | Only Qwen3.5 397B A17B measured |
| Terminal-Bench Hard (Artificial Analysis run) | 40.9% | 11.4% | Qwen3.5 397B A17B |
| ALE-Bench | 621.9 | Not measured | Only Qwen3.5 397B A17B measured |
| LMArena Text: coding | 1465 | 1438 | Qwen3.5 397B A17B |
| LMArena Text: software and IT | 1463 | 1442 | Qwen3.5 397B A17B |
| LMArena WebDev: React | 1388 | Not measured | Only Qwen3.5 397B A17B measured |
| LMArena WebDev: HTML | 1409 | Not measured | Only Qwen3.5 397B A17B measured |
| Neither model has a result on the other 34 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index10
| Benchmark | Qwen3.5 397B A17B | Qwen3 VL 235B A22B Instruct | Leader |
|---|---|---|---|
| AA Intelligence Index | 21.4 | 9.9 | Qwen3.5 397B A17B |
| Epoch Capabilities Index | 146.8 | Not measured | Only Qwen3.5 397B A17B measured |
Agentic workLike-for-likePublic boards, shown as published, kept out of the index10
| Benchmark | Qwen3.5 397B A17B | Qwen3 VL 235B A22B Instruct | Leader |
|---|---|---|---|
| τ²-Bench Telecom | 95.6% | 54.1% | Qwen3.5 397B A17B |
| APEX-Agents | 24.9% | Not measured | Only Qwen3.5 397B A17B measured |
| Neither model has a result on the other 40 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index70
| Benchmark | Qwen3.5 397B A17B | Qwen3 VL 235B A22B Instruct | Leader |
|---|---|---|---|
| GPQA Diamond (Artificial Analysis run) | 89.3% | 77.2% | Qwen3.5 397B A17B |
| Humanity's Last Exam (Artificial Analysis run) | 29.0% | 11.9% | Qwen3.5 397B A17B |
| GPQA Diamond (Epoch AI run) | 86.4% | Not measured | Only Qwen3.5 397B A17B measured |
| Mystery game puzzles | 18.0% | Not measured | Only Qwen3.5 397B A17B measured |
| LMArena Text: hard prompts | 1448 | 1428 | Qwen3.5 397B A17B |
| LMArena Text: expert | 1462 | 1442 | Qwen3.5 397B A17B |
| LMArena Text: hard prompts in English | 1450 | 1439 | Qwen3.5 397B A17B |
| LMArena Vision: diagrams | 1277 | 1255 | Qwen3.5 397B A17B |
| LMArena Vision: homework | 1292 | 1277 | Qwen3.5 397B A17B |
| Neither model has a result on the other 14 boards here | |||
MathLike-for-likePublic boards, shown as published, kept out of the index20
| Benchmark | Qwen3.5 397B A17B | Qwen3 VL 235B A22B Instruct | Leader |
|---|---|---|---|
| AA Math Index | Not measured | 88.3 | Only Qwen3 VL 235B A22B Instruct measured |
| AIME 2025 (Artificial Analysis run) | Not measured | 88.3% | Only Qwen3 VL 235B A22B Instruct measured |
| OTIS Mock AIME | 88.9% | Not measured | Only Qwen3.5 397B A17B measured |
| FrontierMath, tiers 1 to 3 | 31.2% | Not measured | Only Qwen3.5 397B A17B measured |
| LMArena Text: math | 1454 | 1425 | Qwen3.5 397B A17B |
| LMArena Text: mathematical work | 1454 | 1419 | Qwen3.5 397B A17B |
| MathArena | 19.2% | Not measured | Only Qwen3.5 397B A17B measured |
| Neither model has a result on the other 7 boards here | |||
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index40
| Benchmark | Qwen3.5 397B A17B | Qwen3 VL 235B A22B Instruct | Leader |
|---|---|---|---|
| MMLU Pro (Artificial Analysis run) | Not measured | 83.6% | Only Qwen3 VL 235B A22B Instruct measured |
| LMArena Text: science | 1457 | 1445 | Qwen3.5 397B A17B |
| LMArena Text: medicine and healthcare | 1455 | 1431 | Qwen3.5 397B A17B |
| LMArena Text: legal and government | 1441 | 1438 | Qwen3.5 397B A17B |
| LMArena Text: business and finance | 1435 | 1433 | Qwen3.5 397B A17B |
| Neither model has a result on the other 15 boards here | |||
Web designDirectional onlyPublic boards, shown as published, kept out of the index7 boards, none shared
| Benchmark | Qwen3.5 397B A17B | Qwen3 VL 235B A22B Instruct | Leader |
|---|---|---|---|
| Design Arena: websites | 1190 | Not measured | Only Qwen3.5 397B A17B measured |
| Design Arena: UI components | 1161 | Not measured | Only Qwen3.5 397B A17B measured |
| Design Arena: game dev | 1140 | Not measured | Only Qwen3.5 397B A17B measured |
| Design Arena: data viz | 1175 | Not measured | Only Qwen3.5 397B A17B measured |
| Design Arena: 3D | 1162 | Not measured | Only Qwen3.5 397B A17B measured |
| Design Arena: all code categories | 1179 | Not measured | Only Qwen3.5 397B A17B measured |
| Design Arena: SVG | 1139 | Not measured | Only Qwen3.5 397B A17B measured |
| Neither model has a result on the other 2 boards here | |||
GeneralLike-for-likePublic boards, shown as published, kept out of the index252
| Benchmark | Qwen3.5 397B A17B | Qwen3 VL 235B A22B Instruct | Leader |
|---|---|---|---|
| IFBench | 78.8% | 56.5% | Qwen3.5 397B A17B |
| Long context reasoning (AA-LCR) | 77.3% | 63.7% | Qwen3.5 397B A17B |
| DTBench | 87.5% | Not measured | Only Qwen3.5 397B A17B measured |
| LMCA | 37.9% | Not measured | Only Qwen3.5 397B A17B measured |
| LMArena Text | 1438 | 1420 | Qwen3.5 397B A17B |
| LMArena Text: instruction following | 1423 | 1406 | Qwen3.5 397B A17B |
| LMArena Text: creative writing | 1402 | 1366 | Qwen3.5 397B A17B |
| LMArena Vision | 1262 | 1247 | Qwen3.5 397B A17B |
| LMArena Text: English | 1443 | 1429 | Qwen3.5 397B A17B |
| LMArena Text: non-English | 1430 | 1405 | Qwen3.5 397B A17B |
| LMArena Text: Chinese | 1499 | 1462 | Qwen3.5 397B A17B |
| LMArena Text: German | 1444 | 1424 | Qwen3.5 397B A17B |
| LMArena Text: Japanese | 1430 | 1385 | Qwen3.5 397B A17B |
| LMArena Text: Korean | 1385 | 1394 | Qwen3 VL 235B A22B Instruct |
| LMArena Text: French | 1462 | 1452 | Qwen3.5 397B A17B |
| LMArena Text: Spanish | 1441 | 1428 | Qwen3.5 397B A17B |
| LMArena Text: Russian | 1430 | 1408 | Qwen3.5 397B A17B |
| LMArena Text: Polish | 1448 | 1419 | Qwen3.5 397B A17B |
| LMArena Text: multi-turn | 1447 | 1428 | Qwen3.5 397B A17B |
| LMArena Text: longer queries | 1441 | 1420 | Qwen3.5 397B A17B |
| LMArena Text: writing and language | 1416 | 1393 | Qwen3.5 397B A17B |
| LMArena Text: entertainment and media | 1399 | 1368 | Qwen3.5 397B A17B |
| LMArena Vision: English | 1268 | 1256 | Qwen3.5 397B A17B |
| LMArena Vision: Chinese | 1314 | 1283 | Qwen3.5 397B A17B |
| LMArena Vision: reading text | 1271 | 1253 | Qwen3.5 397B A17B |
| LMArena Vision: humor | 1262 | 1209 | Qwen3.5 397B A17B |
| LMArena Vision: captioning | 1230 | 1281 | Qwen3 VL 235B A22B Instruct |
| LMArena Vision: recognizing things | 1265 | 1249 | Qwen3.5 397B A17B |
| LMArena Vision: creative writing | 1260 | 1220 | Qwen3.5 397B A17B |
| Neither model has a result on the other 19 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | Qwen3.5 397B A17B | Qwen3 VL 235B A22B Instruct | 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 VL 235B A22B Instruct costs less
Repository review
50K input + 3K output tokens
Qwen3 VL 235B A22B Instruct costs less
Agent loop
200K input + 10K output tokens
Qwen3 VL 235B A22B Instruct 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 | Qwen3.5 397B A17B | Qwen3 VL 235B A22B Instruct |
|---|---|---|
| Maker | Qwen | Qwen |
| Weights | Open weights | Open weights |
| Released | February 16, 2026 | September 23, 2025 |
| Context window | 262K | 262K |
| Input $/Mtok | $0.39 | $0.21 |
| Output $/Mtok | $2.34 | $1.90 |
| Coding index | Not benchmarked | Not benchmarked |
| Independent measurements behind it | 1 | 0 |
| Cost per solved task | Not computable | Not computable |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is Qwen3.5 397B A17B better than Qwen3 VL 235B A22B Instruct for coding?
Neither model has a coding score yet, so the two cannot be ranked on coding. Pick Qwen3.5 397B A17B for work measured by LMArena Text: Korean or LMArena Vision: captioning, where it leads.
Which is cheaper, Qwen3.5 397B A17B or Qwen3 VL 235B A22B Instruct?
Qwen3.5 397B A17B costs $0.39 per million input tokens and $2.34 per million output tokens; Qwen3 VL 235B A22B Instruct costs $0.21 and $1.90. Cost per solved task is known for neither of them.
Which has the bigger context window, Qwen3.5 397B A17B or Qwen3 VL 235B A22B Instruct?
Neither: both take 262K tokens.
How many benchmarks is the Qwen3.5 397B A17B vs Qwen3 VL 235B A22B Instruct comparison based on?
45 benchmarks have a result for both models as of October 23, 2026. Qwen3.5 397B A17B's coding index rests on 1 independent benchmark and Qwen3 VL 235B A22B Instruct's on 0. Where only one model has a result, the page says so instead of naming a winner.