o1 vs Qwen3 32B
o1 leads Qwen3 32B on the Artificial Analysis Intelligence Index, 15.2 to 7.3, and Qwen3 32B costs 150.0x less per output token ($0.40 vs $60.00 per million). Updated October 22, 2026; 39 benchmarks have a result for both.
o1 or Qwen3 32B: which to pick
Neither model has a coding score yet, so the two cannot be ranked on coding.
Pick o1 for a larger context window (200K against 131K tokens) and work measured by Humanity's Last Exam (Artificial Analysis run) or LMArena Text: expert, where it leads.
- Shared results
- 39
- o1 only
- 30
- Qwen3 32B only
- 6
- 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
o1
o1 leads, 15.2 against 7.3.
Same board, both measured
Reasoning
GPQA Diamond
No call
Only o1 has a GPQA Diamond result, 73.2%.
Directional only
Math
AIME
No call
Only o1 has a AIME result, 71.5%.
Directional only
Knowledge
MMLU Pro
No call
Only o1 has a MMLU Pro result, 83.5%.
Directional only
Long documents
Listed context window
o1
o1 takes 200K tokens against 131K.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
Qwen3 32B
o1 $0.045, Qwen3 32B $0.0003 per turn.
List prices
At a glance
- Coding index
- o1 not benchmarked, Qwen3 32B not benchmarked.
- Widest category
- Agentic work, 32.7 points apart on average over 1 shared benchmark.
- Widest single gap
- Long context reasoning (AA-LCR): o1 65.0% against Qwen3 32B 0.0%, 65.0 points apart on a 0 to 100 scale.
- Output price
- o1 $60.00 against Qwen3 32B $0.40 per million output tokens, 150.0x apart.
- Context window
- o1 takes 200K 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 | o1 | Bars | Qwen3 32B | Ties |
|---|---|---|---|---|
| Coding | 5 | 0 | ||
| Intelligence | 2 | 0 | ||
| Agentic work | 1 | 0 | ||
| Reasoning | 4 | 2 | ||
| Math | 1 | 2 | ||
| Knowledge | 3 | 2 | ||
| General | 16 | 1 | ||
| All shared boards | 32 | 7 |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index50
| Benchmark | o1 | Qwen3 32B | Leader |
|---|---|---|---|
| Aider polyglot | 61.7% | 40.0% | o1 |
| AA Coding Index | 39.7 | Not measured | Only o1 measured |
| LiveCodeBench (Vals AI) | 50.3% | Not measured | Only o1 measured |
| LiveCodeBench (Artificial Analysis run) | 67.9% | 54.6% | o1 |
| SciCode | Not measured | 36.0% | Only Qwen3 32B measured |
| Terminal-Bench Hard (Artificial Analysis run) | 12.9% | 3.0% | o1 |
| WeirdML | 46.1% | Not measured | Only o1 measured |
| LMArena Text: coding | 1367 | 1357 | o1 |
| LMArena Text: software and IT | 1366 | 1366 | o1 |
| SciCode | 4.6% | Not measured | Only o1 measured |
| Neither model has a result on the other 29 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index20
| Benchmark | o1 | Qwen3 32B | Leader |
|---|---|---|---|
| AA Intelligence Index | 15.2 | 7.3 | o1 |
| Epoch Capabilities Index | 141.7 | 138.4 | o1 |
Agentic workLike-for-likePublic boards, shown as published, kept out of the index10
| Benchmark | o1 | Qwen3 32B | Leader |
|---|---|---|---|
| τ²-Bench Telecom | 62.6% | 29.8% | o1 |
| METR time horizons | 51.1% | Not measured | Only o1 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 37 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index42
| Benchmark | o1 | Qwen3 32B | Leader |
|---|---|---|---|
| GPQA Diamond | 73.2% | Not measured | Only o1 measured |
| MMMU Pro (Vals AI) | 77.4% | Not measured | Only o1 measured |
| GPQA Diamond (Artificial Analysis run) | 74.7% | 66.8% | o1 |
| Humanity's Last Exam (Artificial Analysis run) | 7.0% | 7.4% | Qwen3 32B |
| GPQA Diamond (Epoch AI run) | 76.8% | 65.7% | o1 |
| ARC-AGI | 30.7% | Not measured | Only o1 measured |
| SimpleBench | 40.1% | Not measured | Only o1 measured |
| Chess puzzles | 10.0% | Not measured | Only o1 measured |
| VPCT | 37.0% | Not measured | Only o1 measured |
| LMArena Text: hard prompts | 1371 | 1334 | o1 |
| LMArena Text: expert | 1361 | 1362 | Qwen3 32B |
| LMArena Text: hard prompts in English | 1372 | 1350 | o1 |
| Kagi LLM Benchmark | Not measured | 48.7% | Only Qwen3 32B measured |
| Neither model has a result on the other 9 boards here | |||
MathLike-for-likePublic boards, shown as published, kept out of the index12
| Benchmark | o1 | Qwen3 32B | Leader |
|---|---|---|---|
| AIME | 71.5% | Not measured | Only o1 measured |
| MATH 500 (Vals AI) | 90.4% | Not measured | Only o1 measured |
| MGSM (Vals AI) | 89.3% | Not measured | Only o1 measured |
| 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 | 73.3% | 66.9% | o1 |
| FrontierMath, tiers 1 to 3 | 14.7% | Not measured | Only o1 measured |
| MATH level 5 | 94.7% | Not measured | Only o1 measured |
| LMArena Text: math | 1388 | 1399 | Qwen3 32B |
| LMArena Text: mathematical work | 1390 | 1420 | Qwen3 32B |
| Neither model has a result on the other 3 boards here | |||
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index32
| Benchmark | o1 | Qwen3 32B | Leader |
|---|---|---|---|
| MMLU Pro | 83.5% | Not measured | Only o1 measured |
| LegalBench (Vals AI) | 80.4% | Not measured | Only o1 measured |
| TaxEval (Vals AI) | 74.3% | Not measured | Only o1 measured |
| MedQA (Vals AI) | 96.5% | Not measured | Only o1 measured |
| MMLU Pro (Artificial Analysis run) | 84.1% | 79.8% | o1 |
| SimpleQA Verified | 41.1% | Not measured | Only o1 measured |
| GeoBench | 80.0% | Not measured | Only o1 measured |
| LMArena Text: science | 1369 | 1372 | Qwen3 32B |
| LMArena Text: medicine and healthcare | 1355 | 1368 | Qwen3 32B |
| LMArena Text: legal and government | 1368 | 1324 | o1 |
| LMArena Text: business and finance | 1328 | 1318 | o1 |
| Neither model has a result on the other 7 boards here | |||
Web designNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | o1 | Qwen3 32B | Leader |
|---|---|---|---|
| Neither model has a result on 9 boards here | |||
GeneralLike-for-likePublic boards, shown as published, kept out of the index161
| Benchmark | o1 | Qwen3 32B | Leader |
|---|---|---|---|
| IFBench | 70.3% | 36.3% | o1 |
| Long context reasoning (AA-LCR) | 65.0% | 0.0% | o1 |
| DTBench | 74.7% | 67.5% | o1 |
| LMCA | 22.3% | 17.3% | o1 |
| Fiction.LiveBench | 83.3% | 74.2% | o1 |
| Creative writing (Lech Mazur) | 70.2% | Not measured | Only o1 measured |
| LMArena Text | 1366 | 1340 | o1 |
| LMArena Text: instruction following | 1367 | 1305 | o1 |
| LMArena Text: creative writing | 1347 | 1297 | o1 |
| LMArena Vision | 1168 | Not measured | Only o1 measured |
| LMArena Text: English | 1372 | 1359 | o1 |
| LMArena Text: non-English | 1358 | 1317 | o1 |
| LMArena Text: Chinese | 1394 | 1357 | o1 |
| LMArena Text: German | 1337 | 1341 | Qwen3 32B |
| LMArena Text: Japanese | 1346 | Not measured | Only o1 measured |
| LMArena Text: Korean | 1396 | Not measured | Only o1 measured |
| LMArena Text: French | 1344 | Not measured | Only o1 measured |
| LMArena Text: Spanish | 1345 | Not measured | Only o1 measured |
| LMArena Text: Russian | 1356 | 1311 | o1 |
| LMArena Text: multi-turn | 1355 | 1331 | o1 |
| LMArena Text: longer queries | 1378 | 1327 | o1 |
| LMArena Text: writing and language | 1360 | 1298 | o1 |
| LMArena Text: entertainment and media | 1358 | 1279 | o1 |
| LMArena Vision: English | 1162 | Not measured | Only o1 measured |
| LMArena Vision: Chinese | 1235 | Not measured | Only o1 measured |
| Neither model has a result on the other 21 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | o1 | 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 | OpenAI o1 | Qwen3 32B |
|---|---|---|
| Maker | OpenAI | Qwen |
| Weights | No public weights listed | Open weights |
| Released | December 17, 2024 | April 28, 2025 |
| Context window | 200K | 131K |
| Input $/Mtok | $15.00 | $0.14 |
| Output $/Mtok | $60.00 | $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 o1 better than Qwen3 32B for coding?
Neither model has a coding score yet, so the two cannot be ranked on coding. Pick o1 for a larger context window (200K against 131K tokens) and work measured by Humanity's Last Exam (Artificial Analysis run) or LMArena Text: expert, where it leads.
Which is cheaper, o1 or Qwen3 32B?
o1 costs $15.00 per million input tokens and $60.00 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, o1 or Qwen3 32B?
o1, with 200K tokens against 131K.
How many benchmarks is the o1 vs Qwen3 32B comparison based on?
39 benchmarks have a result for both models as of October 22, 2026. o1'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.