Model comparison
Claude 3 Opus vs DeepSeek LLM 67B
Claude 3 Opus is the stronger model overall, scoring 29.5 to 24.9 on the Noometry Index.
Last verified . 15 shared benchmarks.
DeepSeekSummary
- They share 15 benchmarks with published results for both. Claude 3 Opus scores higher in 7 categories and DeepSeek LLM 67B in 1 category; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Claude 3 Opus leads 24.5 to 7.0.
- The biggest single-benchmark swing is MATH Level 5: 37.5% for Claude 3 Opus and 6.4% for DeepSeek LLM 67B.
- DeepSeek LLM 67B has downloadable open weights; the other is API-only.
Side by side
| Claude 3 Opus | DeepSeek LLM 67B | |
|---|---|---|
| Provider | Anthropic | DeepSeek |
| Noometry Index | 29.5 | 24.9 |
| Released | 2024-02-29 | 2023-11-29 |
| Weights | Proprietary | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 46 | 15 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude 3 Opus leads
Claude 3 Opus: 32.9 (#267), DeepSeek LLM 67B: 31.9 (#278)
| Benchmark | Claude 3 Opus | DeepSeek LLM 67B |
|---|---|---|
| LMArena Coding | 1264 | 1096 |
| WeirdML | 19.2% | — |
| BigCodeBench Instruct | 45.5% | — |
| LiveBench Coding | 38.6% | — |
| BigCodeBench Complete | 57.4% | — |
| HumanEval+ | 77.4% | — |
| MBPP+ | 73.3% | — |
Agentic & Tool Use Not comparable
Claude 3 Opus: 24.6 (#116), DeepSeek LLM 67B: —
| Benchmark | Claude 3 Opus | DeepSeek LLM 67B |
|---|---|---|
| Cybench | 10% | — |
| METR Time Horizons | 29.5% | — |
Reasoning DeepSeek LLM 67B leads
Claude 3 Opus: 14.6 (#324), DeepSeek LLM 67B: 16.5 (#304)
| Benchmark | Claude 3 Opus | DeepSeek LLM 67B |
|---|---|---|
| Chess Puzzles | 5% | 0% |
| LMArena Hard Prompts | 1245 | 1070 |
| Epoch Capabilities Index | 126.91 | 110.5 |
| SimpleBench | 23.5% | — |
| EnigmaEval | 0.8% | — |
| LiveBench Reasoning | 40.6% | — |
| DTBench | 61.6% | — |
| LiveBench Data Analysis | 57.9% | — |
| LMCA | 17% | — |
| ForecastBench | 58.4 | — |
| LiveBench | 49.2% | — |
| WinoGrande | 88.5% | — |
Math Claude 3 Opus leads
Claude 3 Opus: 14.8 (#299), DeepSeek LLM 67B: 8.7 (#324)
| Benchmark | Claude 3 Opus | DeepSeek LLM 67B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 4.7% | 0.8% |
| LMArena Math | 1273 | 1108 |
| MATH Level 5 | 37.5% | 6.4% |
| LiveBench Math | 43.6% | — |
Knowledge Claude 3 Opus leads
Claude 3 Opus: 24.5 (#267), DeepSeek LLM 67B: 7.0 (#313)
| Benchmark | Claude 3 Opus | DeepSeek LLM 67B |
|---|---|---|
| GPQA Diamond | 47.2% | 24.6% |
| SimpleQA Verified | 12.6% | — |
| Confabulations | 22.7% | — |
| LMArena Expert | 1223 | — |
| MMLU | 84.6% | — |
Multimodal Not comparable
Claude 3 Opus: 27.1 (#116), DeepSeek LLM 67B: —
| Benchmark | Claude 3 Opus | DeepSeek LLM 67B |
|---|---|---|
| LMArena Vision | 1023 | — |
Multilingual Claude 3 Opus leads
Claude 3 Opus: 41.4 (#207), DeepSeek LLM 67B: 29.4 (#267)
| Benchmark | Claude 3 Opus | DeepSeek LLM 67B |
|---|---|---|
| LMArena Non-English | 1258 | 1073 |
| LMArena Chinese | 1248 | 1132 |
| LMArena French | 1275 | — |
| LMArena German | 1258 | — |
| LMArena Japanese | 1204 | — |
| LMArena Korean | 1187 | — |
| LMArena Russian | 1280 | — |
| LMArena Spanish | 1246 | — |
Instruction Following Claude 3 Opus leads
Claude 3 Opus: 64.1 (#228), DeepSeek LLM 67B: 55.4 (#277)
| Benchmark | Claude 3 Opus | DeepSeek LLM 67B |
|---|---|---|
| LMArena Instruction Following | 1248 | 1079 |
| LiveBench Instruction Following | 63.9% | — |
Long Context Claude 3 Opus leads
Claude 3 Opus: 38.2 (#202), DeepSeek LLM 67B: 33.1 (#265)
| Benchmark | Claude 3 Opus | DeepSeek LLM 67B |
|---|---|---|
| LMArena Longer Query | 1259 | 1092 |
Writing & Preference Claude 3 Opus leads
Claude 3 Opus: 47.2 (#213), DeepSeek LLM 67B: 31.6 (#282)
| Benchmark | Claude 3 Opus | DeepSeek LLM 67B |
|---|---|---|
| LMArena Text | 1262 | 1105 |
| LMArena Creative Writing | 1235 | 1067 |
| LMArena Multi-Turn | 1275 | 1082 |
| LiveBench Language | 50.4% | — |
Frequently asked questions
Is Claude 3 Opus better than DeepSeek LLM 67B?
Claude 3 Opus is the stronger model overall, scoring 29.5 to 24.9 on the Noometry Index.
Is Claude 3 Opus or DeepSeek LLM 67B better for coding?
Claude 3 Opus scores higher on coding benchmarks: 32.9 versus 31.9 in the Noometry coding category.
How many benchmarks do Claude 3 Opus and DeepSeek LLM 67B share?
15 benchmarks have published results for both models. Claude 3 Opus has 46 scored results on Noometry and DeepSeek LLM 67B has 15.