# DeepSeek LLM 67B vs o1

> o1 is the stronger model overall, scoring 40.9 to 24.9 on the Noometry Index.

- Canonical page: https://noometry.com/compare/deepseek-llm-67b-vs-o1
- Last updated: 2026-10-10
- Shared benchmarks: 15

## Summary

- They share 15 benchmarks with published results for both. DeepSeek LLM 67B scores higher in 0 categories and o1 in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where o1 leads 41.5 to 7.0.
- The biggest single-benchmark swing is MATH Level 5: 6.4% for DeepSeek LLM 67B and 94.7% for o1.
- DeepSeek LLM 67B has downloadable open weights; the other is API-only.

## Snapshot

| | DeepSeek LLM 67B | o1 |
|---|---|---|
| Provider | DeepSeek | OpenAI |
| Noometry Index | 24.9 | 40.9 |
| Rank | 347 | 143 |
| Context | — | 200K |
| Input $/M | — | $15 |
| Output $/M | — | $60 |
| Weights | Open | Proprietary |

## Coding

- DeepSeek LLM 67B: 31.9 (#278)
- o1: 46.1 (#70)

| Benchmark | DeepSeek LLM 67B | o1 |
|---|---|---|
| LMArena Coding | 1096 | 1367 |
| Aider Polyglot | — | 61.7% |
| WeirdML | — | 47.6% |
| LiveBench Coding | — | 69.7% |
| CadEval | — | 56% |
| HumanEval+ | — | 89% |
| MBPP+ | — | 80.2% |

## Agentic & Tool Use

- DeepSeek LLM 67B: —
- o1: 24.6 (#117)

| Benchmark | DeepSeek LLM 67B | o1 |
|---|---|---|
| Cybench | — | 10% |
| METR Time Horizons | — | 51.1% |

## Reasoning

- DeepSeek LLM 67B: 16.5 (#304)
- o1: 27.9 (#111)

| Benchmark | DeepSeek LLM 67B | o1 |
|---|---|---|
| Chess Puzzles | 0% | 15% |
| LMArena Hard Prompts | 1070 | 1371 |
| Epoch Capabilities Index | 110.5 | 141.91 |
| SimpleBench | — | 41.7% |
| ARC-AGI-1 | — | 30.7% |
| EnigmaEval | — | 5.7% |
| LiveBench Reasoning | — | 91.6% |
| DTBench | — | 74.7% |
| LiveBench Data Analysis | — | 65.5% |
| LMCA | — | 22.3% |
| LiveBench | — | 75.7% |

## Math

- DeepSeek LLM 67B: 8.7 (#324)
- o1: 36.1 (#175)

| Benchmark | DeepSeek LLM 67B | o1 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 0.8% | 73.3% |
| LMArena Math | 1108 | 1388 |
| MATH Level 5 | 6.4% | 94.7% |
| FrontierMath (Tiers 1-3) | — | 14.7% |
| LiveBench Math | — | 80.3% |
| FrontierMath (Feb 2025 set) | — | 9.3% |

## Knowledge

- DeepSeek LLM 67B: 7.0 (#313)
- o1: 41.5 (#110)

| Benchmark | DeepSeek LLM 67B | o1 |
|---|---|---|
| GPQA Diamond | 24.6% | 76.8% |
| Humanity's Last Exam | — | 8% |
| SimpleQA Verified | — | 41.1% |
| Confabulations | — | 11.7% |
| LMArena Expert | — | 1361 |

## Multimodal

- DeepSeek LLM 67B: —
- o1: 34.2 (#93)

| Benchmark | DeepSeek LLM 67B | o1 |
|---|---|---|
| LMArena Vision | — | 1168 |
| GeoBench | — | 80% |
| VPCT | — | 37% |
| SpatialViz-Bench | — | 41.4% |

## Multilingual

- DeepSeek LLM 67B: 29.4 (#267)
- o1: 48.6 (#142)

| Benchmark | DeepSeek LLM 67B | o1 |
|---|---|---|
| LMArena Non-English | 1073 | 1358 |
| LMArena Chinese | 1132 | 1394 |
| LMArena French | — | 1344 |
| LMArena German | — | 1337 |
| LMArena Japanese | — | 1346 |
| LMArena Korean | — | 1396 |
| LMArena Russian | — | 1356 |
| LMArena Spanish | — | 1345 |

## Instruction Following

- DeepSeek LLM 67B: 55.4 (#277)
- o1: 74.8 (#86)

| Benchmark | DeepSeek LLM 67B | o1 |
|---|---|---|
| LMArena Instruction Following | 1079 | 1367 |
| LiveBench Instruction Following | — | 81.5% |

## Long Context

- DeepSeek LLM 67B: 33.1 (#265)
- o1: 50.3 (#9)

| Benchmark | DeepSeek LLM 67B | o1 |
|---|---|---|
| LMArena Longer Query | 1092 | 1378 |
| Fiction.LiveBench | — | 83.3% |

## Writing & Preference

- DeepSeek LLM 67B: 31.6 (#282)
- o1: 55.6 (#144)

| Benchmark | DeepSeek LLM 67B | o1 |
|---|---|---|
| LMArena Text | 1105 | 1366 |
| LMArena Creative Writing | 1067 | 1348 |
| LMArena Multi-Turn | 1082 | 1369 |
| Short-Story Creative Writing | — | 70.2% |
| LiveBench Language | — | 65.4% |

## FAQ

### Is DeepSeek LLM 67B better than o1?

o1 is the stronger model overall, scoring 40.9 to 24.9 on the Noometry Index.

### Is DeepSeek LLM 67B or o1 better for coding?

o1 scores higher on coding benchmarks: 46.1 versus 31.9 in the Noometry coding category.

### How many benchmarks do DeepSeek LLM 67B and o1 share?

15 benchmarks have published results for both models. DeepSeek LLM 67B has 15 scored results on Noometry and o1 has 52.