Gemini 2.5 Pro vs Qwen3 VL 235B A22B Instruct
Gemini 2.5 Pro leads Qwen3 VL 235B A22B Instruct on the Artificial Analysis Intelligence Index, 16.1 to 9.9, and Qwen3 VL 235B A22B Instruct costs 5.3x less per output token ($1.90 vs $10.00 per million). Updated October 22, 2026; 49 benchmarks have a result for both.
Gemini 2.5 Pro or Qwen3 VL 235B A22B Instruct: which to pick
Only Gemini 2.5 Pro has a coding score, 19.1 on the Vibe Coding Index; Qwen3 VL 235B A22B Instruct has none, so the two cannot be ranked on coding.
Pick Qwen3 VL 235B A22B Instruct for output tokens 5.3x cheaper ($1.90 vs $10.00 per million) and work measured by AA Math Index or AIME 2025 (Artificial Analysis run), where it leads.
- Shared results
- 49
- Gemini 2.5 Pro only
- 41
- Qwen3 VL 235B A22B Instruct only
- 0
- 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
Only Gemini 2.5 Pro has a coding score, so no call is made.
One side not measured
Intelligence
AA Intelligence Index
Gemini 2.5 Pro
Gemini 2.5 Pro leads, 16.1 against 9.9.
Same board, both measured
Agentic work
AA Agentic Index
No call
Only Gemini 2.5 Pro has a AA Agentic Index result, 1.6.
Directional only
Web design
Design Arena: websites
No call
Only Gemini 2.5 Pro has a Design Arena: websites result, 1167.
Directional only
Long documents
Listed context window
Gemini 2.5 Pro
Gemini 2.5 Pro takes 1.05M tokens against 262K.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
Qwen3 VL 235B A22B Instruct
Gemini 2.5 Pro $0.0063, Qwen3 VL 235B A22B Instruct $0.0012 per turn.
List prices
Cost per solved task
One SWE-bench Verified task, solved
No call
Only Gemini 2.5 Pro has the number, $0.64.
Directional only
At a glance
- Coding index
- Gemini 2.5 Pro 19.1 from 3 benchmarks, Qwen3 VL 235B A22B Instruct not benchmarked.
- Widest category
- Coding, 8.9 points apart on average over 4 shared benchmarks.
- Widest single gap
- LiveCodeBench (Artificial Analysis run): Gemini 2.5 Pro 80.1% against Qwen3 VL 235B A22B Instruct 64.6%, 15.5 points apart on a 0 to 100 scale.
- Output price
- Gemini 2.5 Pro $10.00 against Qwen3 VL 235B A22B Instruct $1.90 per million output tokens, 5.3x apart.
- Context window
- Gemini 2.5 Pro takes 1.05M tokens, Qwen3 VL 235B A22B Instruct 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 | Gemini 2.5 Pro | Bars | Qwen3 VL 235B A22B Instruct | Ties |
|---|---|---|---|---|
| Coding | 4 | 0 | ||
| Intelligence | 1 | 0 | ||
| Agentic work | 0 | 0 | 1 tie | |
| Reasoning | 7 | 0 | ||
| Math | 2 | 2 | ||
| Knowledge | 5 | 0 | ||
| General | 26 | 1 | ||
| All shared boards | 45 | 3 | 1 tie |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index40
| Benchmark | Gemini 2.5 Pro | Qwen3 VL 235B A22B Instruct | Leader |
|---|---|---|---|
| SWE-bench Verified | 53.6% | Not measured | Only Gemini 2.5 Pro measured |
| LMArena WebDev | 1226 | Not measured | Only Gemini 2.5 Pro measured |
| AA Coding Index | 33.3 | Not measured | Only Gemini 2.5 Pro measured |
| SWE-bench Verified (Vals AI) | 54.4% | Not measured | Only Gemini 2.5 Pro measured |
| Vibe Code Bench (Vals AI) | 0.4% | Not measured | Only Gemini 2.5 Pro measured |
| LiveCodeBench (Artificial Analysis run) | 80.1% | 64.6% | Gemini 2.5 Pro |
| SciCode | 46.3% | Not measured | Only Gemini 2.5 Pro measured |
| Terminal-Bench Hard (Artificial Analysis run) | 26.5% | 11.4% | Gemini 2.5 Pro |
| Terminal-Bench | 32.6% | Not measured | Only Gemini 2.5 Pro measured |
| GSO-Bench | 3.9% | Not measured | Only Gemini 2.5 Pro measured |
| WeirdML | 54.0% | Not measured | Only Gemini 2.5 Pro measured |
| LMArena Text: coding | 1452 | 1438 | Gemini 2.5 Pro |
| LMArena Text: software and IT | 1459 | 1442 | Gemini 2.5 Pro |
| LMArena WebDev: HTML | 1235 | Not measured | Only Gemini 2.5 Pro measured |
| LMArena WebDev: from an image | 1285 | Not measured | Only Gemini 2.5 Pro measured |
| Neither model has a result on the other 24 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index10
| Benchmark | Gemini 2.5 Pro | Qwen3 VL 235B A22B Instruct | Leader |
|---|---|---|---|
| AA Intelligence Index | 16.1 | 9.9 | Gemini 2.5 Pro |
| Epoch Capabilities Index | 145.1 | Not measured | Only Gemini 2.5 Pro measured |
Agentic workLike-for-likePublic boards, shown as published, kept out of the index00
| Benchmark | Gemini 2.5 Pro | Qwen3 VL 235B A22B Instruct | Leader |
|---|---|---|---|
| AA Agentic Index | 1.6 | Not measured | Only Gemini 2.5 Pro measured |
| τ²-Bench Telecom | 54.1% | 54.1% | Tie |
| GDPval | 23.3% | Not measured | Only Gemini 2.5 Pro measured |
| DeepResearch Bench | 41.5% | Not measured | Only Gemini 2.5 Pro measured |
| Vending-Bench 2 | $574 | Not measured | Only Gemini 2.5 Pro measured |
| BrowseComp-Plus | 29.5% | Not measured | Only Gemini 2.5 Pro measured |
| Neither model has a result on the other 35 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index70
| Benchmark | Gemini 2.5 Pro | Qwen3 VL 235B A22B Instruct | Leader |
|---|---|---|---|
| GPQA Diamond (Artificial Analysis run) | 84.4% | 77.2% | Gemini 2.5 Pro |
| Humanity's Last Exam (Artificial Analysis run) | 22.5% | 11.9% | Gemini 2.5 Pro |
| GPQA Diamond (Epoch AI run) | 85.3% | Not measured | Only Gemini 2.5 Pro measured |
| ARC-AGI | 41.0% | Not measured | Only Gemini 2.5 Pro measured |
| ARC-AGI-2 | 4.9% | Not measured | Only Gemini 2.5 Pro measured |
| LMArena Text: hard prompts | 1455 | 1428 | Gemini 2.5 Pro |
| LMArena Text: expert | 1452 | 1442 | Gemini 2.5 Pro |
| LMArena Text: hard prompts in English | 1453 | 1439 | Gemini 2.5 Pro |
| LMArena Vision: diagrams | 1274 | 1255 | Gemini 2.5 Pro |
| LMArena Vision: homework | 1281 | 1277 | Gemini 2.5 Pro |
| Kagi LLM Benchmark | 70.3% | Not measured | Only Gemini 2.5 Pro measured |
| Neither model has a result on the other 11 boards here | |||
MathLike-for-likePublic boards, shown as published, kept out of the index22
| Benchmark | Gemini 2.5 Pro | Qwen3 VL 235B A22B Instruct | Leader |
|---|---|---|---|
| AA Math Index | 87.7 | 88.3 | Qwen3 VL 235B A22B Instruct |
| AIME 2025 (Artificial Analysis run) | 87.7% | 88.3% | Qwen3 VL 235B A22B Instruct |
| OTIS Mock AIME | 84.7% | Not measured | Only Gemini 2.5 Pro measured |
| FrontierMath, tiers 1 to 3 | 24.6% | Not measured | Only Gemini 2.5 Pro measured |
| FrontierMath, tier 4 | 0.0% | Not measured | Only Gemini 2.5 Pro measured |
| LMArena Text: math | 1450 | 1425 | Gemini 2.5 Pro |
| LMArena Text: mathematical work | 1454 | 1419 | Gemini 2.5 Pro |
| Neither model has a result on the other 6 boards here | |||
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index50
| Benchmark | Gemini 2.5 Pro | Qwen3 VL 235B A22B Instruct | Leader |
|---|---|---|---|
| CaseLaw (Vals AI) | 63.9% | Not measured | Only Gemini 2.5 Pro measured |
| CorpFin (Vals AI) | 60.8% | Not measured | Only Gemini 2.5 Pro measured |
| MortgageTax (Vals AI) | 68.9% | Not measured | Only Gemini 2.5 Pro measured |
| MedCode (Vals AI) | 50.6% | Not measured | Only Gemini 2.5 Pro measured |
| MMLU Pro (Artificial Analysis run) | 86.2% | 83.6% | Gemini 2.5 Pro |
| LMArena Text: science | 1480 | 1444 | Gemini 2.5 Pro |
| LMArena Text: medicine and healthcare | 1476 | 1431 | Gemini 2.5 Pro |
| LMArena Text: legal and government | 1480 | 1438 | Gemini 2.5 Pro |
| LMArena Text: business and finance | 1439 | 1433 | Gemini 2.5 Pro |
| FACTS Benchmark Suite | 61.5% | Not measured | Only Gemini 2.5 Pro measured |
| Neither model has a result on the other 8 boards here | |||
Web designDirectional onlyPublic boards, shown as published, kept out of the index6 boards, none shared
| Benchmark | Gemini 2.5 Pro | Qwen3 VL 235B A22B Instruct | Leader |
|---|---|---|---|
| Design Arena: websites | 1167 | Not measured | Only Gemini 2.5 Pro measured |
| Design Arena: UI components | 1137 | Not measured | Only Gemini 2.5 Pro measured |
| Design Arena: game dev | 1111 | Not measured | Only Gemini 2.5 Pro measured |
| Design Arena: data viz | 1222 | Not measured | Only Gemini 2.5 Pro measured |
| Design Arena: 3D | 1080 | Not measured | Only Gemini 2.5 Pro measured |
| Design Arena: all code categories | 1152 | Not measured | Only Gemini 2.5 Pro measured |
| Neither model has a result on the other 3 boards here | |||
GeneralLike-for-likePublic boards, shown as published, kept out of the index261
| Benchmark | Gemini 2.5 Pro | Qwen3 VL 235B A22B Instruct | Leader |
|---|---|---|---|
| SAGE (Vals AI) | 41.9% | Not measured | Only Gemini 2.5 Pro measured |
| MedScribe (Vals AI) | 73.5% | Not measured | Only Gemini 2.5 Pro measured |
| IFBench | 48.7% | 56.5% | Qwen3 VL 235B A22B Instruct |
| Long context reasoning (AA-LCR) | 69.0% | 63.7% | Gemini 2.5 Pro |
| DTBench | 82.4% | Not measured | Only Gemini 2.5 Pro measured |
| LMCA | 34.8% | Not measured | Only Gemini 2.5 Pro measured |
| Creative writing (Lech Mazur) | 83.8% | Not measured | Only Gemini 2.5 Pro measured |
| LMArena Text | 1458 | 1420 | Gemini 2.5 Pro |
| LMArena Text: instruction following | 1437 | 1406 | Gemini 2.5 Pro |
| LMArena Text: creative writing | 1454 | 1366 | Gemini 2.5 Pro |
| LMArena Vision | 1263 | 1247 | Gemini 2.5 Pro |
| LMArena Text: English | 1458 | 1429 | Gemini 2.5 Pro |
| LMArena Text: non-English | 1452 | 1405 | Gemini 2.5 Pro |
| LMArena Text: Chinese | 1507 | 1463 | Gemini 2.5 Pro |
| LMArena Text: German | 1487 | 1424 | Gemini 2.5 Pro |
| LMArena Text: Japanese | 1461 | 1385 | Gemini 2.5 Pro |
| LMArena Text: Korean | 1434 | 1394 | Gemini 2.5 Pro |
| LMArena Text: French | 1473 | 1452 | Gemini 2.5 Pro |
| LMArena Text: Spanish | 1474 | 1428 | Gemini 2.5 Pro |
| LMArena Text: Russian | 1461 | 1408 | Gemini 2.5 Pro |
| LMArena Text: Polish | 1479 | 1419 | Gemini 2.5 Pro |
| LMArena Text: multi-turn | 1453 | 1428 | Gemini 2.5 Pro |
| LMArena Text: longer queries | 1449 | 1419 | Gemini 2.5 Pro |
| LMArena Text: writing and language | 1454 | 1392 | Gemini 2.5 Pro |
| LMArena Text: entertainment and media | 1441 | 1368 | Gemini 2.5 Pro |
| LMArena Vision: English | 1261 | 1256 | Gemini 2.5 Pro |
| LMArena Vision: Chinese | 1300 | 1283 | Gemini 2.5 Pro |
| LMArena Vision: reading text | 1270 | 1253 | Gemini 2.5 Pro |
| LMArena Vision: humor | 1262 | 1209 | Gemini 2.5 Pro |
| LMArena Vision: captioning | 1304 | 1281 | Gemini 2.5 Pro |
| LMArena Vision: recognizing things | 1283 | 1249 | Gemini 2.5 Pro |
| LMArena Vision: creative writing | 1265 | 1220 | Gemini 2.5 Pro |
| LMArena Document | 1421 | Not measured | Only Gemini 2.5 Pro measured |
| 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 | Gemini 2.5 Pro | 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 | Google Gemini 2.5 Pro | Qwen3 VL 235B A22B Instruct |
|---|---|---|
| Maker | Qwen | |
| Weights | No public weights listed | Open weights |
| Released | June 17, 2025 | September 23, 2025 |
| Context window | 1.05M | 262K |
| Input $/Mtok | $1.25 | $0.21 |
| Output $/Mtok | $10.00 | $1.90 |
| Coding index | 19.1 | Not benchmarked |
| Independent measurements behind it | 3 | 0 |
| Cost per solved task | $0.64 | Not computable |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is Gemini 2.5 Pro better than Qwen3 VL 235B A22B Instruct for coding?
Only Gemini 2.5 Pro has a coding score, 19.1 on the Vibe Coding Index; Qwen3 VL 235B A22B Instruct has none, so the two cannot be ranked on coding. Pick Qwen3 VL 235B A22B Instruct for output tokens 5.3x cheaper ($1.90 vs $10.00 per million) and work measured by AA Math Index or AIME 2025 (Artificial Analysis run), where it leads.
Which is cheaper, Gemini 2.5 Pro or Qwen3 VL 235B A22B Instruct?
Gemini 2.5 Pro costs $1.25 per million input tokens and $10.00 per million output tokens; Qwen3 VL 235B A22B Instruct costs $0.21 and $1.90. Cost per solved task is known for Gemini 2.5 Pro only.
Which has the bigger context window, Gemini 2.5 Pro or Qwen3 VL 235B A22B Instruct?
Gemini 2.5 Pro, with 1.05M tokens against 262K.
How many benchmarks is the Gemini 2.5 Pro vs Qwen3 VL 235B A22B Instruct comparison based on?
49 benchmarks have a result for both models as of October 22, 2026. Gemini 2.5 Pro's coding index rests on 3 independent benchmarks 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.