Qwen2.5 VL 72B Instruct vs Qwen3 32B
Qwen3 32B costs 2.5x less per output token than Qwen2.5 VL 72B Instruct ($0.40 vs $1.00 per million). Updated October 22, 2026; 1 benchmarks have a result for both.
Qwen2.5 VL 72B Instruct or Qwen3 32B: which to pick
Neither model has a coding score yet, so the two cannot be ranked on coding.
Pick Qwen2.5 VL 72B Instruct for work measured by Kagi LLM Benchmark, where it leads.
- Shared results
- 1
- Qwen2.5 VL 72B Instruct only
- 5
- Qwen3 32B only
- 44
- Like-for-like categories
- 1 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 32B has a AA Intelligence Index result, 7.3.
Directional only
Long documents
Listed context window
Qwen3 32B
Qwen3 32B takes 131K tokens against 128K.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
Qwen3 32B
Qwen2.5 VL 72B Instruct $0.0013, Qwen3 32B $0.0003 per turn.
List prices
At a glance
- Coding index
- Qwen2.5 VL 72B Instruct not benchmarked, Qwen3 32B not benchmarked.
- Widest category
- Reasoning, 12.7 points apart on average over 1 shared benchmark.
- Widest single gap
- Kagi LLM Benchmark: Qwen2.5 VL 72B Instruct 36.0% against Qwen3 32B 48.7%, 12.7 points apart on a 0 to 100 scale.
- Output price
- Qwen2.5 VL 72B Instruct $1.00 against Qwen3 32B $0.40 per million output tokens, 2.5x apart.
- Context window
- Qwen2.5 VL 72B Instruct takes 128K tokens, Qwen3 32B 131K.
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 | Qwen2.5 VL 72B Instruct | Bars | Qwen3 32B | Ties |
|---|---|---|---|---|
| Reasoning | 0 | 1 | ||
| All shared boards | 0 | 1 |
CodingDirectional onlySix of these boards feed the Vibe Coding Index6 boards, none shared
| Benchmark | Qwen2.5 VL 72B Instruct | Qwen3 32B | Leader |
|---|---|---|---|
| Aider polyglot | Not measured | 40.0% | Only Qwen3 32B measured |
| LiveCodeBench (Artificial Analysis run) | Not measured | 54.6% | Only Qwen3 32B measured |
| SciCode | Not measured | 36.0% | Only Qwen3 32B measured |
| Terminal-Bench Hard (Artificial Analysis run) | Not measured | 3.0% | Only Qwen3 32B measured |
| LMArena Text: coding | Not measured | 1357 | Only Qwen3 32B measured |
| LMArena Text: software and IT | Not measured | 1366 | Only Qwen3 32B measured |
| Neither model has a result on the other 37 boards here | |||
IntelligenceDirectional onlyPublic boards, shown as published, kept out of the index2 boards, none shared
| Benchmark | Qwen2.5 VL 72B Instruct | Qwen3 32B | Leader |
|---|---|---|---|
| AA Intelligence Index | Not measured | 7.3 | Only Qwen3 32B measured |
| Epoch Capabilities Index | Not measured | 138.4 | Only Qwen3 32B measured |
Agentic workDirectional onlyPublic boards, shown as published, kept out of the index4 boards, none shared
| Benchmark | Qwen2.5 VL 72B Instruct | Qwen3 32B | Leader |
|---|---|---|---|
| τ²-Bench Telecom | Not measured | 29.8% | Only Qwen3 32B measured |
| OSWorld | 5.0% | Not measured | Only Qwen2.5 VL 72B Instruct measured |
| Berkeley Function Calling Leaderboard (BFCL) | Not measured | 48.7% | Only Qwen3 32B measured |
| BrowseComp-Plus | Not measured | 10.7% | Only Qwen3 32B measured |
| Neither model has a result on the other 38 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index01
| Benchmark | Qwen2.5 VL 72B Instruct | Qwen3 32B | Leader |
|---|---|---|---|
| GPQA Diamond (Artificial Analysis run) | Not measured | 66.8% | Only Qwen3 32B measured |
| Humanity's Last Exam (Artificial Analysis run) | Not measured | 7.4% | Only Qwen3 32B measured |
| GPQA Diamond (Epoch AI run) | Not measured | 65.7% | Only Qwen3 32B measured |
| LMArena Text: hard prompts | Not measured | 1334 | Only Qwen3 32B measured |
| LMArena Text: expert | Not measured | 1362 | Only Qwen3 32B measured |
| LMArena Text: hard prompts in English | Not measured | 1350 | Only Qwen3 32B measured |
| Kagi LLM Benchmark | 36.0% | 48.7% | Qwen3 32B |
| Neither model has a result on the other 16 boards here | |||
MathDirectional onlyPublic boards, shown as published, kept out of the index5 boards, none shared
| Benchmark | Qwen2.5 VL 72B Instruct | Qwen3 32B | Leader |
|---|---|---|---|
| AA Math Index | Not measured | 73.0 | Only Qwen3 32B measured |
| AIME 2025 (Artificial Analysis run) | Not measured | 73.0% | Only Qwen3 32B measured |
| OTIS Mock AIME | Not measured | 66.9% | Only Qwen3 32B measured |
| LMArena Text: math | Not measured | 1399 | Only Qwen3 32B measured |
| LMArena Text: mathematical work | Not measured | 1420 | Only Qwen3 32B measured |
| Neither model has a result on the other 9 boards here | |||
KnowledgeDirectional onlyPublic boards, shown as published, kept out of the index6 boards, none shared
| Benchmark | Qwen2.5 VL 72B Instruct | Qwen3 32B | Leader |
|---|---|---|---|
| MMLU Pro (Artificial Analysis run) | Not measured | 79.8% | Only Qwen3 32B measured |
| GeoBench | 62.0% | Not measured | Only Qwen2.5 VL 72B Instruct measured |
| LMArena Text: science | Not measured | 1372 | Only Qwen3 32B measured |
| LMArena Text: medicine and healthcare | Not measured | 1368 | Only Qwen3 32B measured |
| LMArena Text: legal and government | Not measured | 1324 | Only Qwen3 32B measured |
| LMArena Text: business and finance | Not measured | 1318 | Only Qwen3 32B measured |
| Neither model has a result on the other 14 boards here | |||
Web designNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | Qwen2.5 VL 72B Instruct | Qwen3 32B | Leader |
|---|---|---|---|
| Neither model has a result on 9 boards here | |||
GeneralDirectional onlyPublic boards, shown as published, kept out of the index20 boards, none shared
| Benchmark | Qwen2.5 VL 72B Instruct | Qwen3 32B | Leader |
|---|---|---|---|
| IFBench | Not measured | 36.3% | Only Qwen3 32B measured |
| Long context reasoning (AA-LCR) | Not measured | 0.0% | Only Qwen3 32B measured |
| DTBench | Not measured | 67.5% | Only Qwen3 32B measured |
| LMCA | Not measured | 17.3% | Only Qwen3 32B measured |
| Fiction.LiveBench | Not measured | 74.2% | Only Qwen3 32B measured |
| LMArena Text | Not measured | 1340 | Only Qwen3 32B measured |
| LMArena Text: instruction following | Not measured | 1305 | Only Qwen3 32B measured |
| LMArena Text: creative writing | Not measured | 1297 | Only Qwen3 32B measured |
| LMArena Vision | 1107 | Not measured | Only Qwen2.5 VL 72B Instruct measured |
| LMArena Text: English | Not measured | 1359 | Only Qwen3 32B measured |
| LMArena Text: non-English | Not measured | 1317 | Only Qwen3 32B measured |
| LMArena Text: Chinese | Not measured | 1357 | Only Qwen3 32B measured |
| LMArena Text: German | Not measured | 1341 | Only Qwen3 32B measured |
| LMArena Text: Russian | Not measured | 1311 | Only Qwen3 32B measured |
| LMArena Text: multi-turn | Not measured | 1331 | Only Qwen3 32B measured |
| LMArena Text: longer queries | Not measured | 1327 | Only Qwen3 32B measured |
| LMArena Text: writing and language | Not measured | 1298 | Only Qwen3 32B measured |
| LMArena Text: entertainment and media | Not measured | 1279 | Only Qwen3 32B measured |
| LMArena Vision: English | 1115 | Not measured | Only Qwen2.5 VL 72B Instruct measured |
| LMArena Vision: Chinese | 1107 | Not measured | Only Qwen2.5 VL 72B Instruct measured |
| Neither model has a result on the other 28 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | Qwen2.5 VL 72B Instruct | Qwen3 32B | 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 32B costs less
Repository review
50K input + 3K output tokens
Qwen3 32B costs less
Agent loop
200K input + 10K output tokens
Qwen3 32B 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 | Qwen2.5 VL 72B Instruct | Qwen3 32B |
|---|---|---|
| Maker | Qwen | Qwen |
| Weights | Open weights | Open weights |
| Released | February 1, 2025 | April 28, 2025 |
| Context window | 128K | 131K |
| Input $/Mtok | $0.80 | $0.14 |
| Output $/Mtok | $1.00 | $0.40 |
| 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 Qwen2.5 VL 72B Instruct better than Qwen3 32B for coding?
Neither model has a coding score yet, so the two cannot be ranked on coding. Pick Qwen2.5 VL 72B Instruct for work measured by Kagi LLM Benchmark, where it leads.
Which is cheaper, Qwen2.5 VL 72B Instruct or Qwen3 32B?
Qwen2.5 VL 72B Instruct costs $0.80 per million input tokens and $1.00 per million output tokens; Qwen3 32B costs $0.14 and $0.40. Cost per solved task is known for neither of them.
Which has the bigger context window, Qwen2.5 VL 72B Instruct or Qwen3 32B?
Qwen3 32B, with 131K tokens against 128K.
How many benchmarks is the Qwen2.5 VL 72B Instruct vs Qwen3 32B comparison based on?
1 benchmarks have a result for both models as of October 22, 2026. Qwen2.5 VL 72B Instruct's coding index rests on 0 independent benchmarks and Qwen3 32B's on 0. Where only one model has a result, the page says so instead of naming a winner.