Model comparison
Claude 3 Opus vs DeepSeek-V3.2-Exp
DeepSeek-V3.2-Exp is the stronger model overall, scoring 44.3 to 29.5 on the Noometry Index.
Last verified . 24 shared benchmarks.
DeepSeekSummary
- They share 24 benchmarks with published results for both. Claude 3 Opus scores higher in 0 categories and DeepSeek-V3.2-Exp in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where DeepSeek-V3.2-Exp leads 51.7 to 24.5.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 4.7% for Claude 3 Opus and 87.8% for DeepSeek-V3.2-Exp.
- DeepSeek-V3.2-Exp has downloadable open weights; the other is API-only.
Side by side
| Claude 3 Opus | DeepSeek-V3.2-Exp | |
|---|---|---|
| Provider | Anthropic | DeepSeek |
| Noometry Index | 29.5 | 44.3 |
| Released | 2024-02-29 | 2025-09-29 |
| Weights | Proprietary | Open |
| Context window | — | 164K |
| Max output | — | 66K |
| Input $ / M tokens | — | $0.26 |
| Output $ / M tokens | — | $0.38 |
| Results tracked | 46 | 49 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek-V3.2-Exp leads
Claude 3 Opus: 32.9 (#267), DeepSeek-V3.2-Exp: 46.5 (#65)
| Benchmark | Claude 3 Opus | DeepSeek-V3.2-Exp |
|---|---|---|
| WeirdML | 19.2% | 39.5% |
| LMArena Coding | 1264 | 1454 |
| SWE-bench Verified (bash only) | — | 70% |
| Aider Polyglot | — | 74.2% |
| LMArena WebDev | — | 1362 |
| SWE-bench Multilingual | — | 59% |
| SciCode | — | 38.9% |
| BigCodeBench Instruct | 45.5% | — |
| LiveBench Coding | 38.6% | — |
| BigCodeBench Complete | 57.4% | — |
| HumanEval+ | 77.4% | — |
| MBPP+ | 73.3% | — |
Agentic & Tool Use DeepSeek-V3.2-Exp leads
Claude 3 Opus: 24.6 (#116), DeepSeek-V3.2-Exp: 32.7 (#59)
| Benchmark | Claude 3 Opus | DeepSeek-V3.2-Exp |
|---|---|---|
| Terminal-Bench | — | 39.6% |
| APEX-Agents | — | 21.3% |
| Berkeley Function Calling Leaderboard | — | 56.7% |
| TheAgentCompany | — | 42.9% |
| Cybench | 10% | — |
| METR Time Horizons | 29.5% | — |
| Vending-Bench 2 | — | 1,034 |
Reasoning DeepSeek-V3.2-Exp leads
Claude 3 Opus: 14.6 (#324), DeepSeek-V3.2-Exp: 22.1 (#208)
| Benchmark | Claude 3 Opus | DeepSeek-V3.2-Exp |
|---|---|---|
| Chess Puzzles | 5% | 14% |
| LMArena Hard Prompts | 1245 | 1434 |
| DTBench | 61.6% | 87.7% |
| LMCA | 17% | 29.1% |
| Epoch Capabilities Index | 126.91 | 146.27 |
| ARC-AGI-2 | — | 4% |
| SimpleBench | 23.5% | — |
| Kagi LLM Benchmark | — | 52.2% |
| NYT Connections (extended) | — | 36.7% |
| ARC-AGI-1 | — | 57% |
| CritPt | — | 2.9% |
| EnigmaEval | 0.8% | — |
| Thematic Generalization | — | 65% |
| LiveBench Reasoning | 40.6% | — |
| LiveBench Data Analysis | 57.9% | — |
| ForecastBench | 58.4 | — |
| LiveBench | 49.2% | — |
| WinoGrande | 88.5% | — |
Math DeepSeek-V3.2-Exp leads
Claude 3 Opus: 14.8 (#299), DeepSeek-V3.2-Exp: 41.7 (#87)
| Benchmark | Claude 3 Opus | DeepSeek-V3.2-Exp |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 4.7% | 87.8% |
| LMArena Math | 1273 | 1435 |
| MathArena Final-Answer Competitions | — | 57.7% |
| ProofBench | — | 8% |
| LiveBench Math | 43.6% | — |
| MATH Level 5 | 37.5% | — |
| FrontierMath (Feb 2025 set) | — | 22.1% |
| FrontierMath Tier 4 (v1) | — | 2.1% |
Knowledge DeepSeek-V3.2-Exp leads
Claude 3 Opus: 24.5 (#267), DeepSeek-V3.2-Exp: 51.7 (#66)
| Benchmark | Claude 3 Opus | DeepSeek-V3.2-Exp |
|---|---|---|
| GPQA Diamond | 47.2% | 83.4% |
| LMArena Expert | 1223 | 1436 |
| SimpleQA Verified | 12.6% | — |
| Confabulations | 22.7% | — |
| Vectara Hallucination Rate | — | 5.3% |
| MMLU | 84.6% | — |
Multimodal Not comparable
Claude 3 Opus: 27.1 (#116), DeepSeek-V3.2-Exp: —
| Benchmark | Claude 3 Opus | DeepSeek-V3.2-Exp |
|---|---|---|
| LMArena Vision | 1023 | — |
Multilingual DeepSeek-V3.2-Exp leads
Claude 3 Opus: 41.4 (#207), DeepSeek-V3.2-Exp: 52.2 (#90)
| Benchmark | Claude 3 Opus | DeepSeek-V3.2-Exp |
|---|---|---|
| LMArena Non-English | 1258 | 1409 |
| LMArena Chinese | 1248 | 1461 |
| LMArena French | 1275 | 1433 |
| LMArena German | 1258 | 1440 |
| LMArena Japanese | 1204 | 1374 |
| LMArena Korean | 1187 | 1371 |
| LMArena Russian | 1280 | 1424 |
| LMArena Spanish | 1246 | 1440 |
Instruction Following DeepSeek-V3.2-Exp leads
Claude 3 Opus: 64.1 (#228), DeepSeek-V3.2-Exp: 74.5 (#93)
| Benchmark | Claude 3 Opus | DeepSeek-V3.2-Exp |
|---|---|---|
| LMArena Instruction Following | 1248 | 1413 |
| LiveBench Instruction Following | 63.9% | — |
Long Context DeepSeek-V3.2-Exp leads
Claude 3 Opus: 38.2 (#202), DeepSeek-V3.2-Exp: 47.6 (#16)
| Benchmark | Claude 3 Opus | DeepSeek-V3.2-Exp |
|---|---|---|
| LMArena Longer Query | 1259 | 1428 |
| Fiction.LiveBench | — | 83.3% |
| CL-bench | — | 13.2% |
| CL-bench Life | — | 9.5% |
Writing & Preference DeepSeek-V3.2-Exp leads
Claude 3 Opus: 47.2 (#213), DeepSeek-V3.2-Exp: 62.4 (#77)
| Benchmark | Claude 3 Opus | DeepSeek-V3.2-Exp |
|---|---|---|
| LMArena Text | 1262 | 1425 |
| LMArena Creative Writing | 1235 | 1403 |
| LMArena Multi-Turn | 1275 | 1427 |
| EQ-Bench Creative Writing | — | 1515 |
| LiveBench Language | 50.4% | — |
Frequently asked questions
Is Claude 3 Opus better than DeepSeek-V3.2-Exp?
DeepSeek-V3.2-Exp is the stronger model overall, scoring 44.3 to 29.5 on the Noometry Index.
Is Claude 3 Opus or DeepSeek-V3.2-Exp better for coding?
DeepSeek-V3.2-Exp scores higher on coding benchmarks: 46.5 versus 32.9 in the Noometry coding category.
How many benchmarks do Claude 3 Opus and DeepSeek-V3.2-Exp share?
24 benchmarks have published results for both models. Claude 3 Opus has 46 scored results on Noometry and DeepSeek-V3.2-Exp has 49.