Qwen3 30B A3B Instruct 2507 vs Qwen3.5 Plus 2026-02-15
Qwen3 30B A3B Instruct 2507 costs 5.2x less per output token than Qwen3.5 Plus 2026-02-15 ($0.30 vs $1.56 per million). Updated October 23, 2026; no benchmark has tested both models yet.
- Coding
- No data
- Overall
- 41.5
- Output $/M
- $0.30
- Coding
- No data
- Overall
- No data
- Output $/M
- $1.56
Qwen3 30B A3B Instruct 2507 or Qwen3.5 Plus 2026-02-15: which to pick
Neither model has a coding score yet, so the two cannot be ranked on coding.
Pick Qwen3 30B A3B Instruct 2507 for output tokens 5.2x cheaper ($0.30 vs $1.56 per million).
No benchmark has tested both yet. Qwen3 30B A3B Instruct 2507 has results on LiveCodeBench (Artificial Analysis run), SciCode and Terminal-Bench Hard (Artificial Analysis run); Qwen3.5 Plus 2026-02-15 on Design Arena: websites, Design Arena: UI components and Design Arena: game dev. Compare them on price and context instead.
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 30B A3B Instruct 2507 has a AA Intelligence Index result, 7.5.
Directional only
Web design
Design Arena: websites
No call
Only Qwen3.5 Plus 2026-02-15 has a Design Arena: websites result, 1187.
Directional only
Long documents
Listed context window
Qwen3.5 Plus 2026-02-15
Qwen3.5 Plus 2026-02-15 takes 1M tokens against 262K.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
Qwen3 30B A3B Instruct 2507
Qwen3 30B A3B Instruct 2507 $0.0002, Qwen3.5 Plus 2026-02-15 $0.0010 per turn.
List prices
At a glance
- Coding index
- Qwen3 30B A3B Instruct 2507 not benchmarked, Qwen3.5 Plus 2026-02-15 not benchmarked.
- Widest category
- None yet: no category has been tested on both models.
- Widest single gap
- None yet: no benchmark has tested both models.
- Output price
- Qwen3 30B A3B Instruct 2507 $0.30 against Qwen3.5 Plus 2026-02-15 $1.56 per million output tokens, 5.2x apart.
- Context window
- Qwen3 30B A3B Instruct 2507 takes 262K tokens, Qwen3.5 Plus 2026-02-15 1M.
Benchmark by benchmark
Every board either model has a result on, a category to a fold; bars are each result on a 0 to 100 scale. Only one model is measured on each, so no row names a leader.
CodingDirectional onlySix of these boards feed the Vibe Coding Index5 boards measured
| Benchmark | Qwen3 30B A3B Instruct 2507 | Qwen3.5 Plus 2026-02-15 | Leader |
|---|---|---|---|
| LiveCodeBench (Artificial Analysis run) | 70.7% | Not measured | Only Qwen3 30B A3B Instruct 2507 measured |
| SciCode | 33.0% | Not measured | Only Qwen3 30B A3B Instruct 2507 measured |
| Terminal-Bench Hard (Artificial Analysis run) | 6.1% | Not measured | Only Qwen3 30B A3B Instruct 2507 measured |
| LMArena Text: coding | 1416 | Not measured | Only Qwen3 30B A3B Instruct 2507 measured |
| LMArena Text: software and IT | 1417 | Not measured | Only Qwen3 30B A3B Instruct 2507 measured |
| Neither model has a result on the other 38 boards here | |||
IntelligenceDirectional onlyPublic boards, shown as published, kept out of the index1 board measured
| Benchmark | Qwen3 30B A3B Instruct 2507 | Qwen3.5 Plus 2026-02-15 | Leader |
|---|---|---|---|
| AA Intelligence Index | 7.5 | Not measured | Only Qwen3 30B A3B Instruct 2507 measured |
| Neither model has a result on the other 1 board here | |||
Agentic workDirectional onlyPublic boards, shown as published, kept out of the index2 boards measured
| Benchmark | Qwen3 30B A3B Instruct 2507 | Qwen3.5 Plus 2026-02-15 | Leader |
|---|---|---|---|
| τ²-Bench Telecom | 28.1% | Not measured | Only Qwen3 30B A3B Instruct 2507 measured |
| Berkeley Function Calling Leaderboard (BFCL) | 41.4% | Not measured | Only Qwen3 30B A3B Instruct 2507 measured |
| Neither model has a result on the other 40 boards here | |||
ReasoningDirectional onlyPublic boards, shown as published, kept out of the index6 boards measured
| Benchmark | Qwen3 30B A3B Instruct 2507 | Qwen3.5 Plus 2026-02-15 | Leader |
|---|---|---|---|
| GPQA Diamond (Artificial Analysis run) | 70.7% | Not measured | Only Qwen3 30B A3B Instruct 2507 measured |
| Humanity's Last Exam (Artificial Analysis run) | 10.3% | Not measured | Only Qwen3 30B A3B Instruct 2507 measured |
| GPQA Diamond (Epoch AI run) | 55.6% | Not measured | Only Qwen3 30B A3B Instruct 2507 measured |
| LMArena Text: hard prompts | 1398 | Not measured | Only Qwen3 30B A3B Instruct 2507 measured |
| LMArena Text: expert | 1396 | Not measured | Only Qwen3 30B A3B Instruct 2507 measured |
| LMArena Text: hard prompts in English | 1406 | Not measured | Only Qwen3 30B A3B Instruct 2507 measured |
| Neither model has a result on the other 17 boards here | |||
MathDirectional onlyPublic boards, shown as published, kept out of the index5 boards measured
| Benchmark | Qwen3 30B A3B Instruct 2507 | Qwen3.5 Plus 2026-02-15 | Leader |
|---|---|---|---|
| AA Math Index | 66.3 | Not measured | Only Qwen3 30B A3B Instruct 2507 measured |
| AIME 2025 (Artificial Analysis run) | 66.3% | Not measured | Only Qwen3 30B A3B Instruct 2507 measured |
| OTIS Mock AIME | 62.2% | Not measured | Only Qwen3 30B A3B Instruct 2507 measured |
| LMArena Text: math | 1395 | Not measured | Only Qwen3 30B A3B Instruct 2507 measured |
| LMArena Text: mathematical work | 1412 | Not measured | Only Qwen3 30B A3B Instruct 2507 measured |
| Neither model has a result on the other 9 boards here | |||
KnowledgeDirectional onlyPublic boards, shown as published, kept out of the index5 boards measured
| Benchmark | Qwen3 30B A3B Instruct 2507 | Qwen3.5 Plus 2026-02-15 | Leader |
|---|---|---|---|
| MMLU Pro (Artificial Analysis run) | 80.5% | Not measured | Only Qwen3 30B A3B Instruct 2507 measured |
| LMArena Text: science | 1393 | Not measured | Only Qwen3 30B A3B Instruct 2507 measured |
| LMArena Text: medicine and healthcare | 1392 | Not measured | Only Qwen3 30B A3B Instruct 2507 measured |
| LMArena Text: legal and government | 1387 | Not measured | Only Qwen3 30B A3B Instruct 2507 measured |
| LMArena Text: business and finance | 1389 | Not measured | Only Qwen3 30B A3B Instruct 2507 measured |
| Neither model has a result on the other 15 boards here | |||
Web designDirectional onlyPublic boards, shown as published, kept out of the index8 boards measured
| Benchmark | Qwen3 30B A3B Instruct 2507 | Qwen3.5 Plus 2026-02-15 | Leader |
|---|---|---|---|
| Design Arena: websites | Not measured | 1187 | Only Qwen3.5 Plus 2026-02-15 measured |
| Design Arena: UI components | Not measured | 1175 | Only Qwen3.5 Plus 2026-02-15 measured |
| Design Arena: game dev | Not measured | 1105 | Only Qwen3.5 Plus 2026-02-15 measured |
| Design Arena: data viz | Not measured | 1135 | Only Qwen3.5 Plus 2026-02-15 measured |
| Design Arena: 3D | Not measured | 1120 | Only Qwen3.5 Plus 2026-02-15 measured |
| Design Arena: all code categories | Not measured | 1166 | Only Qwen3.5 Plus 2026-02-15 measured |
| Design Arena: SVG | Not measured | 1107 | Only Qwen3.5 Plus 2026-02-15 measured |
| Design Arena: ASCII art | Not measured | 1094 | Only Qwen3.5 Plus 2026-02-15 measured |
| Neither model has a result on the other 1 board here | |||
GeneralDirectional onlyPublic boards, shown as published, kept out of the index21 boards measured
| Benchmark | Qwen3 30B A3B Instruct 2507 | Qwen3.5 Plus 2026-02-15 | Leader |
|---|---|---|---|
| IFBench | 50.7% | Not measured | Only Qwen3 30B A3B Instruct 2507 measured |
| Long context reasoning (AA-LCR) | 61.3% | Not measured | Only Qwen3 30B A3B Instruct 2507 measured |
| DTBench | 67.2% | Not measured | Only Qwen3 30B A3B Instruct 2507 measured |
| LMCA | 22.4% | Not measured | Only Qwen3 30B A3B Instruct 2507 measured |
| LMArena Text | 1384 | Not measured | Only Qwen3 30B A3B Instruct 2507 measured |
| LMArena Text: instruction following | 1363 | Not measured | Only Qwen3 30B A3B Instruct 2507 measured |
| LMArena Text: creative writing | 1317 | Not measured | Only Qwen3 30B A3B Instruct 2507 measured |
| LMArena Text: English | 1391 | Not measured | Only Qwen3 30B A3B Instruct 2507 measured |
| LMArena Text: non-English | 1372 | Not measured | Only Qwen3 30B A3B Instruct 2507 measured |
| LMArena Text: Chinese | 1433 | Not measured | Only Qwen3 30B A3B Instruct 2507 measured |
| LMArena Text: German | 1379 | Not measured | Only Qwen3 30B A3B Instruct 2507 measured |
| LMArena Text: Japanese | 1338 | Not measured | Only Qwen3 30B A3B Instruct 2507 measured |
| LMArena Text: Korean | 1331 | Not measured | Only Qwen3 30B A3B Instruct 2507 measured |
| LMArena Text: French | 1419 | Not measured | Only Qwen3 30B A3B Instruct 2507 measured |
| LMArena Text: Spanish | 1404 | Not measured | Only Qwen3 30B A3B Instruct 2507 measured |
| LMArena Text: Russian | 1369 | Not measured | Only Qwen3 30B A3B Instruct 2507 measured |
| LMArena Text: Polish | 1353 | Not measured | Only Qwen3 30B A3B Instruct 2507 measured |
| LMArena Text: multi-turn | 1378 | Not measured | Only Qwen3 30B A3B Instruct 2507 measured |
| LMArena Text: longer queries | 1378 | Not measured | Only Qwen3 30B A3B Instruct 2507 measured |
| LMArena Text: writing and language | 1336 | Not measured | Only Qwen3 30B A3B Instruct 2507 measured |
| LMArena Text: entertainment and media | 1334 | Not measured | Only Qwen3 30B A3B Instruct 2507 measured |
| Neither model has a result on the other 27 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 30B A3B Instruct 2507 costs less
Repository review
50K input + 3K output tokens
Qwen3 30B A3B Instruct 2507 costs less
Agent loop
200K input + 10K output tokens
Qwen3 30B A3B Instruct 2507 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 30B A3B Instruct 2507 | Qwen3.5 Plus 2026-02-15 |
|---|---|---|
| Maker | Qwen | Qwen |
| Weights | Open weights | No public weights listed |
| Released | July 29, 2025 | February 16, 2026 |
| Context window | 262K | 1M |
| Input $/Mtok | $0.09 | $0.26 |
| Output $/Mtok | $0.30 | $1.56 |
| Coding index | Not benchmarked | Not benchmarked |
| Independent measurements behind it | 0 | 0 |
| Cost per solved task | Not computable | Not computable |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is Qwen3 30B A3B Instruct 2507 better than Qwen3.5 Plus 2026-02-15 for coding?
Neither model has a coding score yet, so the two cannot be ranked on coding. Pick Qwen3 30B A3B Instruct 2507 for output tokens 5.2x cheaper ($0.30 vs $1.56 per million).
Which is cheaper, Qwen3 30B A3B Instruct 2507 or Qwen3.5 Plus 2026-02-15?
Qwen3 30B A3B Instruct 2507 costs $0.09 per million input tokens and $0.30 per million output tokens; Qwen3.5 Plus 2026-02-15 costs $0.26 and $1.56. Cost per solved task is known for neither of them.
Which has the bigger context window, Qwen3 30B A3B Instruct 2507 or Qwen3.5 Plus 2026-02-15?
Qwen3.5 Plus 2026-02-15, with 1M tokens against 262K.
Has any benchmark tested both Qwen3 30B A3B Instruct 2507 and Qwen3.5 Plus 2026-02-15?
No benchmark has tested both yet. Qwen3 30B A3B Instruct 2507 has results on LiveCodeBench (Artificial Analysis run), SciCode and Terminal-Bench Hard (Artificial Analysis run); Qwen3.5 Plus 2026-02-15 on Design Arena: websites, Design Arena: UI components and Design Arena: game dev. Compare them on price and context instead. Data as of October 23, 2026.