Model comparison
DeepSeek-V3.2-Exp vs o1-mini
DeepSeek-V3.2-Exp is the stronger model overall, scoring 44.9 to 33.7 on the Noometry Index.
Last verified . 25 shared benchmarks.
DeepSeekSummary
- They share 25 benchmarks with published results for both. DeepSeek-V3.2-Exp scores higher in 9 categories and o1-mini in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where DeepSeek-V3.2-Exp leads 51.6 to 34.8.
- The biggest single-benchmark swing is ARC-AGI-1: 57% for DeepSeek-V3.2-Exp and 14% for o1-mini.
- DeepSeek-V3.2-Exp has downloadable open weights; the other is API-only.
Side by side
| DeepSeek-V3.2-Exp | o1-mini | |
|---|---|---|
| Provider | DeepSeek | OpenAI |
| Noometry Index | 44.9 | 33.7 |
| Released | 2025-09-29 | 2024-09-12 |
| Weights | Open | Proprietary |
| Context window | 164K | — |
| Max output | 66K | — |
| Input $ / M tokens | $0.26 | — |
| Output $ / M tokens | $0.38 | — |
| Results tracked | 48 | 39 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek-V3.2-Exp leads
DeepSeek-V3.2-Exp: 46.4 (#61), o1-mini: 35.1 (#218)
| Benchmark | DeepSeek-V3.2-Exp | o1-mini |
|---|---|---|
| Aider Polyglot | 74.2% | 32.9% |
| WeirdML | 39.5% | 36.3% |
| LMArena Coding | 1454 | 1362 |
| SWE-bench Verified (bash only) | 70% | — |
| LMArena WebDev | 1362 | — |
| SWE-bench Multilingual | 59% | — |
| SciCode | 38.9% | — |
| LiveBench Coding | — | 48% |
| HumanEval+ | — | 89% |
| MBPP+ | — | 78.8% |
Agentic & Tool Use DeepSeek-V3.2-Exp leads
DeepSeek-V3.2-Exp: 37.9 (#31), o1-mini: 24.3 (#112)
| Benchmark | DeepSeek-V3.2-Exp | o1-mini |
|---|---|---|
| Terminal-Bench | 39.6% | — |
| Berkeley Function Calling Leaderboard | 56.7% | — |
| TheAgentCompany | 42.9% | — |
| Cybench | — | 10% |
| Vending-Bench 2 | 1,034 | — |
Reasoning DeepSeek-V3.2-Exp leads
DeepSeek-V3.2-Exp: 22.0 (#185), o1-mini: 8.9 (#340)
| Benchmark | DeepSeek-V3.2-Exp | o1-mini |
|---|---|---|
| ARC-AGI-2 | 4% | 0.8% |
| ARC-AGI-1 | 57% | 14% |
| LMArena Hard Prompts | 1433 | 1333 |
| Epoch Capabilities Index | 146.28 | 135.82 |
| SimpleBench | — | 18.1% |
| Kagi LLM Benchmark | 52.2% | — |
| NYT Connections (extended) | 36.7% | — |
| CritPt | 2.9% | — |
| Chess Puzzles | 14% | — |
| Thematic Generalization | 65% | — |
| LiveBench Reasoning | — | 72.3% |
| DTBench | 87.7% | — |
| LiveBench Data Analysis | — | 57.9% |
| LMCA | 29.1% | — |
| LiveBench | — | 57.8% |
Math DeepSeek-V3.2-Exp leads
DeepSeek-V3.2-Exp: 40.8 (#75), o1-mini: 33.7 (#168)
| Benchmark | DeepSeek-V3.2-Exp | o1-mini |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 87.8% | 46.9% |
| LMArena Math | 1435 | 1358 |
| FrontierMath (Feb 2025 set) | 22.1% | 1.7% |
| MathArena Final-Answer Competitions | 57.7% | — |
| ProofBench | 8% | — |
| LiveBench Math | — | 62% |
| MATH Level 5 | — | 89.2% |
| FrontierMath Tier 4 (v1) | 2.1% | — |
Knowledge DeepSeek-V3.2-Exp leads
DeepSeek-V3.2-Exp: 51.6 (#61), o1-mini: 34.8 (#180)
| Benchmark | DeepSeek-V3.2-Exp | o1-mini |
|---|---|---|
| GPQA Diamond | 83.4% | 62.4% |
| LMArena Expert | 1436 | 1316 |
| Confabulations | — | 18.6% |
| Vectara Hallucination Rate | 5.3% | — |
Multilingual DeepSeek-V3.2-Exp leads
DeepSeek-V3.2-Exp: 52.4 (#85), o1-mini: 43.8 (#177)
| Benchmark | DeepSeek-V3.2-Exp | o1-mini |
|---|---|---|
| LMArena Non-English | 1409 | 1289 |
| LMArena Chinese | 1461 | 1314 |
| LMArena French | 1435 | 1292 |
| LMArena German | 1440 | 1278 |
| LMArena Japanese | 1377 | 1245 |
| LMArena Korean | 1372 | 1223 |
| LMArena Russian | 1423 | 1283 |
| LMArena Spanish | 1441 | 1302 |
Instruction Following DeepSeek-V3.2-Exp leads
DeepSeek-V3.2-Exp: 74.7 (#89), o1-mini: 67.0 (#200)
| Benchmark | DeepSeek-V3.2-Exp | o1-mini |
|---|---|---|
| LMArena Instruction Following | 1413 | 1304 |
| LiveBench Instruction Following | — | 65.4% |
Long Context DeepSeek-V3.2-Exp leads
DeepSeek-V3.2-Exp: 47.6 (#15), o1-mini: 40.1 (#156)
| Benchmark | DeepSeek-V3.2-Exp | o1-mini |
|---|---|---|
| LMArena Longer Query | 1428 | 1321 |
| Fiction.LiveBench | 83.3% | — |
| CL-bench | 13.2% | — |
| CL-bench Life | 9.5% | — |
Writing & Preference DeepSeek-V3.2-Exp leads
DeepSeek-V3.2-Exp: 62.5 (#73), o1-mini: 48.6 (#196)
| Benchmark | DeepSeek-V3.2-Exp | o1-mini |
|---|---|---|
| LMArena Text | 1425 | 1317 |
| LMArena Creative Writing | 1403 | 1244 |
| LMArena Multi-Turn | 1427 | 1314 |
| Short-Story Creative Writing | — | 64.9% |
| EQ-Bench Creative Writing | 1515 | — |
| LiveBench Language | — | 40.9% |
Frequently asked questions
Is DeepSeek-V3.2-Exp better than o1-mini?
DeepSeek-V3.2-Exp is the stronger model overall, scoring 44.9 to 33.7 on the Noometry Index.
Is DeepSeek-V3.2-Exp or o1-mini better for coding?
DeepSeek-V3.2-Exp scores higher on coding benchmarks: 46.4 versus 35.1 in the Noometry coding category.
How many benchmarks do DeepSeek-V3.2-Exp and o1-mini share?
25 benchmarks have published results for both models. DeepSeek-V3.2-Exp has 48 scored results on Noometry and o1-mini has 39.