Model comparison
DeepSeek-V3.1 vs o1
DeepSeek-V3.1 is the stronger model overall, scoring 42.8 to 40.9 on the Noometry Index.
Last verified . 23 shared benchmarks.
DeepSeekSummary
- They share 23 benchmarks with published results for both. DeepSeek-V3.1 scores higher in 5 categories and o1 in 3 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in long context, where o1 leads 50.3 to 36.3.
- The biggest single-benchmark swing is Fiction.LiveBench: 52.8% for DeepSeek-V3.1 and 83.3% for o1.
- DeepSeek-V3.1 is cheaper at $0.25 / $0.95 per million input/output tokens, against $15 / $60 for o1.
- o1 accepts more context: 200K tokens versus 164K.
- DeepSeek-V3.1 has downloadable open weights; the other is API-only.
Side by side
| DeepSeek-V3.1 | o1 | |
|---|---|---|
| Provider | DeepSeek | OpenAI |
| Noometry Index | 42.8 | 40.9 |
| Released | 2025-08-21 | 2024-09-12 |
| Weights | Open | Proprietary |
| Context window | 164K | 200K |
| Max output | 8K | 100K |
| Input $ / M tokens | $0.25 | $15 |
| Output $ / M tokens | $0.95 | $60 |
| Results tracked | 27 | 52 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding o1 leads
DeepSeek-V3.1: 40.3 (#144), o1: 46.1 (#70)
| Benchmark | DeepSeek-V3.1 | o1 |
|---|---|---|
| WeirdML | 38.4% | 47.6% |
| LMArena Coding | 1417 | 1367 |
| Aider Polyglot | — | 61.7% |
| LiveBench Coding | — | 69.7% |
| CadEval | — | 56% |
| HumanEval+ | — | 89% |
| MBPP+ | — | 80.2% |
Agentic & Tool Use Not comparable
DeepSeek-V3.1: —, o1: 24.6 (#117)
| Benchmark | DeepSeek-V3.1 | o1 |
|---|---|---|
| Cybench | — | 10% |
| METR Time Horizons | — | 51.1% |
Reasoning Too close to call
DeepSeek-V3.1: 27.9 (#110), o1: 27.9 (#111)
| Benchmark | DeepSeek-V3.1 | o1 |
|---|---|---|
| SimpleBench | 40% | 41.7% |
| LMArena Hard Prompts | 1417 | 1371 |
| DTBench | 82.7% | 74.7% |
| LMCA | 24.3% | 22.3% |
| Epoch Capabilities Index | 139.92 | 141.91 |
| Kagi LLM Benchmark | 53.2% | — |
| ARC-AGI-1 | — | 30.7% |
| Chess Puzzles | — | 15% |
| EnigmaEval | — | 5.7% |
| LiveBench Reasoning | — | 91.6% |
| LiveBench Data Analysis | — | 65.5% |
| ForecastBench | 58 | — |
| LiveBench | — | 75.7% |
Math DeepSeek-V3.1 leads
DeepSeek-V3.1: 38.9 (#122), o1: 36.1 (#175)
| Benchmark | DeepSeek-V3.1 | o1 |
|---|---|---|
| LMArena Math | 1420 | 1388 |
| FrontierMath (Tiers 1-3) | — | 14.7% |
| OTIS Mock AIME 2024-2025 | — | 73.3% |
| LiveBench Math | — | 80.3% |
| MATH Level 5 | — | 94.7% |
| FrontierMath (Feb 2025 set) | — | 9.3% |
Knowledge DeepSeek-V3.1 leads
DeepSeek-V3.1: 43.7 (#90), o1: 41.5 (#110)
| Benchmark | DeepSeek-V3.1 | o1 |
|---|---|---|
| LMArena Expert | 1405 | 1361 |
| GPQA Diamond | — | 76.8% |
| Humanity's Last Exam | — | 8% |
| SimpleQA Verified | — | 41.1% |
| Confabulations | — | 11.7% |
| Vectara Hallucination Rate | 5.5% | — |
Multimodal Not comparable
DeepSeek-V3.1: —, o1: 34.2 (#93)
| Benchmark | DeepSeek-V3.1 | o1 |
|---|---|---|
| LMArena Vision | — | 1168 |
| GeoBench | — | 80% |
| VPCT | — | 37% |
| SpatialViz-Bench | — | 41.4% |
Multilingual DeepSeek-V3.1 leads
DeepSeek-V3.1: 51.6 (#106), o1: 48.6 (#142)
| Benchmark | DeepSeek-V3.1 | o1 |
|---|---|---|
| LMArena Non-English | 1400 | 1358 |
| LMArena Chinese | 1469 | 1394 |
| LMArena French | 1447 | 1344 |
| LMArena German | 1411 | 1337 |
| LMArena Japanese | 1378 | 1346 |
| LMArena Korean | 1337 | 1396 |
| LMArena Russian | 1405 | 1356 |
| LMArena Spanish | 1431 | 1345 |
Instruction Following Too close to call
DeepSeek-V3.1: 73.9 (#110), o1: 74.8 (#86)
| Benchmark | DeepSeek-V3.1 | o1 |
|---|---|---|
| LMArena Instruction Following | 1400 | 1367 |
| LiveBench Instruction Following | — | 81.5% |
Long Context o1 leads
DeepSeek-V3.1: 36.3 (#232), o1: 50.3 (#9)
| Benchmark | DeepSeek-V3.1 | o1 |
|---|---|---|
| Fiction.LiveBench | 52.8% | 83.3% |
| LMArena Longer Query | 1422 | 1378 |
Writing & Preference DeepSeek-V3.1 leads
DeepSeek-V3.1: 60.3 (#98), o1: 55.6 (#144)
| Benchmark | DeepSeek-V3.1 | o1 |
|---|---|---|
| LMArena Text | 1420 | 1366 |
| LMArena Creative Writing | 1401 | 1348 |
| LMArena Multi-Turn | 1408 | 1369 |
| Short-Story Creative Writing | — | 70.2% |
| EQ-Bench Creative Writing | 1436 | — |
| LiveBench Language | — | 65.4% |
Frequently asked questions
Is DeepSeek-V3.1 better than o1?
DeepSeek-V3.1 is the stronger model overall, scoring 42.8 to 40.9 on the Noometry Index.
Which is cheaper, DeepSeek-V3.1 or o1?
DeepSeek-V3.1 is cheaper. It lists at $0.25 per million input tokens and $0.95 per million output tokens; o1 lists at $15 and $60.
Is DeepSeek-V3.1 or o1 better for coding?
o1 scores higher on coding benchmarks: 46.1 versus 40.3 in the Noometry coding category.
Which has the bigger context window?
o1 does, with 200K tokens against 164K.
How many benchmarks do DeepSeek-V3.1 and o1 share?
23 benchmarks have published results for both models. DeepSeek-V3.1 has 27 scored results on Noometry and o1 has 52.