Qwen3 Coder 480B A35B vs Hy3
Hy3 costs 4.5x less per output token than Qwen3 Coder 480B A35B ($0.33 vs $1.50 per million). Updated October 21, 2026; 40 benchmarks have a result for both.
Qwen3 Coder 480B A35B or Hy3: which to pick
Neither model has a coding score yet, so the two cannot be ranked on coding.
Pick Qwen3 Coder 480B A35B for work measured by LMArena WebDev or ALE-Bench, where it leads.
- Shared results
- 40
- Qwen3 Coder 480B A35B only
- 13
- Hy3 only
- 4
- Like-for-like categories
- 6 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
No call
Only Qwen3 Coder 480B A35B has a AA Intelligence Index result, 11.9.
Directional only
Web design
Design Arena: websites
Hy3
Hy3 leads, 1230 against 1159.
Same board, both measured
Long documents
Listed context window
No call
Both take 262K tokens.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
Hy3
Qwen3 Coder 480B A35B $0.0010, Hy3 $0.0002 per turn.
List prices
At a glance
- Coding index
- Qwen3 Coder 480B A35B not benchmarked, Hy3 not benchmarked.
- Widest category
- Coding, 23.6 points apart on average over 6 shared benchmarks.
- Widest single gap
- LMArena WebDev: React: Qwen3 Coder 480B A35B 1257 against Hy3 1511, 42.2 points apart on a 0 to 100 scale.
- Output price
- Qwen3 Coder 480B A35B $1.50 against Hy3 $0.33 per million output tokens, 4.5x 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 Coder 480B A35B | Bars | Hy3 | Ties |
|---|---|---|---|---|
| Coding | 0 | 6 | ||
| Reasoning | 0 | 5 | ||
| Math | 0 | 2 | ||
| Knowledge | 0 | 4 | ||
| Web design | 0 | 5 | ||
| General | 0 | 18 | ||
| All shared boards | 0 | 40 |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index06
| Benchmark | Qwen3 Coder 480B A35B | Hy3 | Leader |
|---|---|---|---|
| SWE-bench Verified | 69.6% | Not measured | Only Qwen3 Coder 480B A35B measured |
| LMArena WebDev | 1275 | 1507 | Hy3 |
| LiveCodeBench (Artificial Analysis run) | 58.5% | Not measured | Only Qwen3 Coder 480B A35B measured |
| SciCode | Not measured | 48.6% | Only Hy3 measured |
| Terminal-Bench Hard (Artificial Analysis run) | 18.9% | Not measured | Only Qwen3 Coder 480B A35B measured |
| Terminal-Bench | 27.2% | Not measured | Only Qwen3 Coder 480B A35B measured |
| GSO-Bench | 4.9% | Not measured | Only Qwen3 Coder 480B A35B measured |
| WeirdML | 41.2% | Not measured | Only Qwen3 Coder 480B A35B measured |
| ALE-Bench | 461.4 | 838.6 | Hy3 |
| LMArena Text: coding | 1412 | 1466 | Hy3 |
| LMArena Text: software and IT | 1393 | 1463 | Hy3 |
| LMArena WebDev: React | 1257 | 1511 | Hy3 |
| LMArena WebDev: HTML | 1287 | 1487 | Hy3 |
| Neither model has a result on the other 26 boards here | |||
IntelligenceDirectional onlyPublic boards, shown as published, kept out of the index1 board, none shared
| Benchmark | Qwen3 Coder 480B A35B | Hy3 | Leader |
|---|---|---|---|
| AA Intelligence Index | 11.9 | Not measured | Only Qwen3 Coder 480B A35B measured |
| Neither model has a result on the other 1 board here | |||
Agentic workDirectional onlyPublic boards, shown as published, kept out of the index1 board, none shared
| Benchmark | Qwen3 Coder 480B A35B | Hy3 | Leader |
|---|---|---|---|
| τ²-Bench Telecom | 43.6% | Not measured | Only Qwen3 Coder 480B A35B measured |
| Neither model has a result on the other 40 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index05
| Benchmark | Qwen3 Coder 480B A35B | Hy3 | Leader |
|---|---|---|---|
| GPQA Diamond (Artificial Analysis run) | 61.8% | 89.7% | Hy3 |
| Humanity's Last Exam (Artificial Analysis run) | 4.5% | 33.5% | Hy3 |
| LMArena Text: hard prompts | 1372 | 1447 | Hy3 |
| LMArena Text: expert | 1338 | 1461 | Hy3 |
| LMArena Text: hard prompts in English | 1380 | 1455 | Hy3 |
| Kagi LLM Benchmark | 49.5% | Not measured | Only Qwen3 Coder 480B A35B measured |
| Neither model has a result on the other 16 boards here | |||
MathLike-for-likePublic boards, shown as published, kept out of the index02
| Benchmark | Qwen3 Coder 480B A35B | Hy3 | Leader |
|---|---|---|---|
| AA Math Index | 39.3 | Not measured | Only Qwen3 Coder 480B A35B measured |
| AIME 2025 (Artificial Analysis run) | 39.3% | Not measured | Only Qwen3 Coder 480B A35B measured |
| LMArena Text: math | 1365 | 1473 | Hy3 |
| LMArena Text: mathematical work | 1367 | 1482 | Hy3 |
| Neither model has a result on the other 9 boards here | |||
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index04
| Benchmark | Qwen3 Coder 480B A35B | Hy3 | Leader |
|---|---|---|---|
| MMLU Pro (Artificial Analysis run) | 78.8% | Not measured | Only Qwen3 Coder 480B A35B measured |
| LMArena Text: science | 1347 | 1457 | Hy3 |
| LMArena Text: medicine and healthcare | 1347 | 1452 | Hy3 |
| LMArena Text: legal and government | 1358 | 1422 | Hy3 |
| LMArena Text: business and finance | 1343 | 1424 | Hy3 |
| Neither model has a result on the other 13 boards here | |||
Web designLike-for-likePublic boards, shown as published, kept out of the index05
| Benchmark | Qwen3 Coder 480B A35B | Hy3 | Leader |
|---|---|---|---|
| Design Arena: websites | 1159 | 1230 | Hy3 |
| Design Arena: UI components | 1113 | 1207 | Hy3 |
| Design Arena: game dev | 1099 | 1208 | Hy3 |
| Design Arena: data viz | 1084 | 1163 | Hy3 |
| Design Arena: 3D | Not measured | 1233 | Only Hy3 measured |
| Design Arena: all code categories | 1143 | 1227 | Hy3 |
| Design Arena: SVG | Not measured | 1200 | Only Hy3 measured |
| Design Arena: ASCII art | Not measured | 1152 | Only Hy3 measured |
| Neither model has a result on the other 1 board here | |||
GeneralLike-for-likePublic boards, shown as published, kept out of the index018
| Benchmark | Qwen3 Coder 480B A35B | Hy3 | Leader |
|---|---|---|---|
| IFBench | 40.5% | Not measured | Only Qwen3 Coder 480B A35B measured |
| Long context reasoning (AA-LCR) | 45.7% | 79.0% | Hy3 |
| LMArena Text | 1357 | 1438 | Hy3 |
| LMArena Text: instruction following | 1355 | 1426 | Hy3 |
| LMArena Text: creative writing | 1333 | 1403 | Hy3 |
| LMArena Text: English | 1359 | 1448 | Hy3 |
| LMArena Text: non-English | 1346 | 1426 | Hy3 |
| LMArena Text: Chinese | 1357 | 1485 | Hy3 |
| LMArena Text: German | 1325 | 1449 | Hy3 |
| LMArena Text: Japanese | 1311 | 1397 | Hy3 |
| LMArena Text: Korean | 1305 | 1396 | Hy3 |
| LMArena Text: French | 1398 | 1456 | Hy3 |
| LMArena Text: Spanish | 1360 | 1457 | Hy3 |
| LMArena Text: Russian | 1366 | 1432 | Hy3 |
| LMArena Text: Polish | 1344 | 1447 | Hy3 |
| LMArena Text: multi-turn | 1366 | 1436 | Hy3 |
| LMArena Text: longer queries | 1378 | 1442 | Hy3 |
| LMArena Text: writing and language | 1340 | 1414 | Hy3 |
| LMArena Text: entertainment and media | 1317 | 1400 | Hy3 |
| Neither model has a result on the other 27 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | Qwen3 Coder 480B A35B | Hy3 | 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
Hy3 costs less
Repository review
50K input + 3K output tokens
Hy3 costs less
Agent loop
200K input + 10K output tokens
Hy3 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 Coder 480B A35B | Tencent Hy3 |
|---|---|---|
| Maker | Qwen | Tencent |
| Weights | Open weights | Open weights |
| Released | July 23, 2025 | July 6, 2026 |
| Context window | 262K | 262K |
| Input $/Mtok | $0.30 | $0.0825 |
| Output $/Mtok | $1.50 | $0.33 |
| Coding index | Not benchmarked | Not benchmarked |
| Independent measurements behind it | 1 | 1 |
| Cost per solved task | Not computable | Not computable |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is Qwen3 Coder 480B A35B better than Hy3 for coding?
Neither model has a coding score yet, so the two cannot be ranked on coding. Pick Qwen3 Coder 480B A35B for work measured by LMArena WebDev or ALE-Bench, where it leads.
Which is cheaper, Qwen3 Coder 480B A35B or Hy3?
Qwen3 Coder 480B A35B costs $0.30 per million input tokens and $1.50 per million output tokens; Hy3 costs $0.0825 and $0.33. Cost per solved task is known for neither of them.
Which has the bigger context window, Qwen3 Coder 480B A35B or Hy3?
Neither: both take 262K tokens.
How many benchmarks is the Qwen3 Coder 480B A35B vs Hy3 comparison based on?
40 benchmarks have a result for both models as of October 21, 2026. Qwen3 Coder 480B A35B's coding index rests on 1 independent benchmark and Hy3's on 1. Where only one model has a result, the page says so instead of naming a winner.