Model comparison
DeepSeek-V3.2-Exp vs o1
DeepSeek-V3.2-Exp is the stronger model overall, scoring 44.3 to 40.9 on the Noometry Index.
Last verified . 28 shared benchmarks.
DeepSeekSummary
- They share 28 benchmarks with published results for both. DeepSeek-V3.2-Exp scores higher in 6 categories and o1 in 3 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where DeepSeek-V3.2-Exp leads 51.7 to 41.5.
- The biggest single-benchmark swing is ARC-AGI-1: 57% for DeepSeek-V3.2-Exp and 30.7% for o1.
- DeepSeek-V3.2-Exp is cheaper at $0.26 / $0.38 per million input/output tokens, against $15 / $60 for o1.
- o1 accepts more context: 200K tokens versus 164K.
- DeepSeek-V3.2-Exp has downloadable open weights; the other is API-only.
Side by side
| DeepSeek-V3.2-Exp | o1 | |
|---|---|---|
| Provider | DeepSeek | OpenAI |
| Noometry Index | 44.3 | 40.9 |
| Released | 2025-09-29 | 2024-09-12 |
| Weights | Open | Proprietary |
| Context window | 164K | 200K |
| Max output | 66K | 100K |
| Input $ / M tokens | $0.26 | $15 |
| Output $ / M tokens | $0.38 | $60 |
| Results tracked | 49 | 52 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
DeepSeek-V3.2-Exp: 46.5 (#65), o1: 46.1 (#70)
| Benchmark | DeepSeek-V3.2-Exp | o1 |
|---|---|---|
| Aider Polyglot | 74.2% | 61.7% |
| WeirdML | 39.5% | 47.6% |
| LMArena Coding | 1454 | 1367 |
| SWE-bench Verified (bash only) | 70% | — |
| LMArena WebDev | 1362 | — |
| SWE-bench Multilingual | 59% | — |
| SciCode | 38.9% | — |
| LiveBench Coding | — | 69.7% |
| CadEval | — | 56% |
| HumanEval+ | — | 89% |
| MBPP+ | — | 80.2% |
Agentic & Tool Use DeepSeek-V3.2-Exp leads
DeepSeek-V3.2-Exp: 32.7 (#59), o1: 24.6 (#117)
| Benchmark | DeepSeek-V3.2-Exp | o1 |
|---|---|---|
| Terminal-Bench | 39.6% | — |
| APEX-Agents | 21.3% | — |
| Berkeley Function Calling Leaderboard | 56.7% | — |
| TheAgentCompany | 42.9% | — |
| Cybench | — | 10% |
| METR Time Horizons | — | 51.1% |
| Vending-Bench 2 | 1,034 | — |
Reasoning o1 leads
DeepSeek-V3.2-Exp: 22.1 (#208), o1: 27.9 (#111)
| Benchmark | DeepSeek-V3.2-Exp | o1 |
|---|---|---|
| ARC-AGI-1 | 57% | 30.7% |
| Chess Puzzles | 14% | 15% |
| LMArena Hard Prompts | 1434 | 1371 |
| DTBench | 87.7% | 74.7% |
| LMCA | 29.1% | 22.3% |
| Epoch Capabilities Index | 146.27 | 141.91 |
| ARC-AGI-2 | 4% | — |
| SimpleBench | — | 41.7% |
| Kagi LLM Benchmark | 52.2% | — |
| NYT Connections (extended) | 36.7% | — |
| CritPt | 2.9% | — |
| EnigmaEval | — | 5.7% |
| Thematic Generalization | 65% | — |
| LiveBench Reasoning | — | 91.6% |
| LiveBench Data Analysis | — | 65.5% |
| LiveBench | — | 75.7% |
Math DeepSeek-V3.2-Exp leads
DeepSeek-V3.2-Exp: 41.7 (#87), o1: 36.1 (#175)
| Benchmark | DeepSeek-V3.2-Exp | o1 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 87.8% | 73.3% |
| LMArena Math | 1435 | 1388 |
| FrontierMath (Feb 2025 set) | 22.1% | 9.3% |
| FrontierMath (Tiers 1-3) | — | 14.7% |
| MathArena Final-Answer Competitions | 57.7% | — |
| ProofBench | 8% | — |
| LiveBench Math | — | 80.3% |
| MATH Level 5 | — | 94.7% |
| FrontierMath Tier 4 (v1) | 2.1% | — |
Knowledge DeepSeek-V3.2-Exp leads
DeepSeek-V3.2-Exp: 51.7 (#66), o1: 41.5 (#110)
| Benchmark | DeepSeek-V3.2-Exp | o1 |
|---|---|---|
| GPQA Diamond | 83.4% | 76.8% |
| LMArena Expert | 1436 | 1361 |
| Humanity's Last Exam | — | 8% |
| SimpleQA Verified | — | 41.1% |
| Confabulations | — | 11.7% |
| Vectara Hallucination Rate | 5.3% | — |
Multimodal Not comparable
DeepSeek-V3.2-Exp: —, o1: 34.2 (#93)
| Benchmark | DeepSeek-V3.2-Exp | o1 |
|---|---|---|
| LMArena Vision | — | 1168 |
| GeoBench | — | 80% |
| VPCT | — | 37% |
| SpatialViz-Bench | — | 41.4% |
Multilingual DeepSeek-V3.2-Exp leads
DeepSeek-V3.2-Exp: 52.2 (#90), o1: 48.6 (#142)
| Benchmark | DeepSeek-V3.2-Exp | o1 |
|---|---|---|
| LMArena Non-English | 1409 | 1358 |
| LMArena Chinese | 1461 | 1394 |
| LMArena French | 1433 | 1344 |
| LMArena German | 1440 | 1337 |
| LMArena Japanese | 1374 | 1346 |
| LMArena Korean | 1371 | 1396 |
| LMArena Russian | 1424 | 1356 |
| LMArena Spanish | 1440 | 1345 |
Instruction Following Too close to call
DeepSeek-V3.2-Exp: 74.5 (#93), o1: 74.8 (#86)
| Benchmark | DeepSeek-V3.2-Exp | o1 |
|---|---|---|
| LMArena Instruction Following | 1413 | 1367 |
| LiveBench Instruction Following | — | 81.5% |
Long Context o1 leads
DeepSeek-V3.2-Exp: 47.6 (#16), o1: 50.3 (#9)
| Benchmark | DeepSeek-V3.2-Exp | o1 |
|---|---|---|
| Fiction.LiveBench | 83.3% | 83.3% |
| LMArena Longer Query | 1428 | 1378 |
| CL-bench | 13.2% | — |
| CL-bench Life | 9.5% | — |
Writing & Preference DeepSeek-V3.2-Exp leads
DeepSeek-V3.2-Exp: 62.4 (#77), o1: 55.6 (#144)
| Benchmark | DeepSeek-V3.2-Exp | o1 |
|---|---|---|
| LMArena Text | 1425 | 1366 |
| LMArena Creative Writing | 1403 | 1348 |
| LMArena Multi-Turn | 1427 | 1369 |
| Short-Story Creative Writing | — | 70.2% |
| EQ-Bench Creative Writing | 1515 | — |
| LiveBench Language | — | 65.4% |
Frequently asked questions
Is DeepSeek-V3.2-Exp better than o1?
DeepSeek-V3.2-Exp is the stronger model overall, scoring 44.3 to 40.9 on the Noometry Index.
Which is cheaper, DeepSeek-V3.2-Exp or o1?
DeepSeek-V3.2-Exp is cheaper. It lists at $0.26 per million input tokens and $0.38 per million output tokens; o1 lists at $15 and $60.
Is DeepSeek-V3.2-Exp or o1 better for coding?
They score almost the same on coding (46.5 vs 46.1); test both on your own repository before choosing.
Which has the bigger context window?
o1 does, with 200K tokens against 164K.
How many benchmarks do DeepSeek-V3.2-Exp and o1 share?
28 benchmarks have published results for both models. DeepSeek-V3.2-Exp has 49 scored results on Noometry and o1 has 52.