Model comparison
Claude 3 Opus vs Qwen2.5 72B Instruct
Qwen2.5 72B Instruct is the stronger model overall, scoring 31.9 to 29.5 on the Noometry Index.
Last verified . 31 shared benchmarks.
Alibaba (Qwen)Summary
- They share 31 benchmarks with published results for both. Claude 3 Opus scores higher in 3 categories and Qwen2.5 72B Instruct in 6 categories; 5 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Qwen2.5 72B Instruct leads 22.3 to 14.6.
- The biggest single-benchmark swing is MATH Level 5: 37.5% for Claude 3 Opus and 63.2% for Qwen2.5 72B Instruct.
- Qwen2.5 72B Instruct has downloadable open weights; the other is API-only.
Side by side
| Claude 3 Opus | Qwen2.5 72B Instruct | |
|---|---|---|
| Provider | Anthropic | Alibaba (Qwen) |
| Noometry Index | 29.5 | 31.9 |
| Released | 2024-02-29 | 2024-09 |
| Weights | Proprietary | Open |
| Context window | — | 131K |
| Max output | — | 8K |
| Input $ / M tokens | — | $1.40 |
| Output $ / M tokens | — | $5.60 |
| Results tracked | 46 | 43 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
Claude 3 Opus: 32.9 (#267), Qwen2.5 72B Instruct: 33.2 (#260)
| Benchmark | Claude 3 Opus | Qwen2.5 72B Instruct |
|---|---|---|
| WeirdML | 19.2% | 16% |
| BigCodeBench Instruct | 45.5% | 45.8% |
| LMArena Coding | 1264 | 1292 |
| BigCodeBench Complete | 57.4% | 55.9% |
| LiveBench Coding | 38.6% | — |
| HumanEval+ | 77.4% | — |
| MBPP+ | 73.3% | — |
Agentic & Tool Use Claude 3 Opus leads
Claude 3 Opus: 24.6 (#116), Qwen2.5 72B Instruct: 22.1 (#133)
| Benchmark | Claude 3 Opus | Qwen2.5 72B Instruct |
|---|---|---|
| METR Time Horizons | 29.5% | 35.8% |
| TheAgentCompany | — | 5.7% |
| Cybench | 10% | — |
| BALROG | — | 16.2% |
Reasoning Qwen2.5 72B Instruct leads
Claude 3 Opus: 14.6 (#324), Qwen2.5 72B Instruct: 22.3 (#199)
| Benchmark | Claude 3 Opus | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Hard Prompts | 1245 | 1271 |
| DTBench | 61.6% | 62.9% |
| LMCA | 17% | 13.4% |
| Epoch Capabilities Index | 126.91 | 129 |
| ForecastBench | 58.4 | 57.5 |
| WinoGrande | 88.5% | 82.3% |
| SimpleBench | 23.5% | — |
| Chess Puzzles | 5% | — |
| EnigmaEval | 0.8% | — |
| LiveBench Reasoning | 40.6% | — |
| LiveBench Data Analysis | 57.9% | — |
| BIG-Bench Hard | — | 79.8% |
| HellaSwag | — | 84.8% |
| LiveBench | 49.2% | — |
| PIQA | — | 82.6% |
Math Qwen2.5 72B Instruct leads
Claude 3 Opus: 14.8 (#299), Qwen2.5 72B Instruct: 19.3 (#287)
| Benchmark | Claude 3 Opus | Qwen2.5 72B Instruct |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 4.7% | 8.1% |
| LMArena Math | 1273 | 1283 |
| MATH Level 5 | 37.5% | 63.2% |
| Omni-MATH | — | 33% |
| LiveBench Math | 43.6% | — |
Knowledge Qwen2.5 72B Instruct leads
Claude 3 Opus: 24.5 (#267), Qwen2.5 72B Instruct: 27.0 (#253)
| Benchmark | Claude 3 Opus | Qwen2.5 72B Instruct |
|---|---|---|
| GPQA Diamond | 47.2% | 49.1% |
| Confabulations | 22.7% | 19.1% |
| LMArena Expert | 1223 | 1245 |
| MMLU | 84.6% | 85.3% |
| SimpleQA Verified | 12.6% | — |
| MMLU-Pro | — | 63.1% |
| GPQA (HELM) | — | 42.6% |
| ARC (AI2) Challenge | — | 94.5% |
| TriviaQA | — | 71.9% |
Multimodal Not comparable
Claude 3 Opus: 27.1 (#116), Qwen2.5 72B Instruct: —
| Benchmark | Claude 3 Opus | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Vision | 1023 | — |
Multilingual Too close to call
Claude 3 Opus: 41.4 (#207), Qwen2.5 72B Instruct: 41.0 (#213)
| Benchmark | Claude 3 Opus | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Non-English | 1258 | 1252 |
| LMArena Chinese | 1248 | 1272 |
| LMArena French | 1275 | 1280 |
| LMArena German | 1258 | 1234 |
| LMArena Japanese | 1204 | 1180 |
| LMArena Korean | 1187 | 1188 |
| LMArena Russian | 1280 | 1264 |
| LMArena Spanish | 1246 | 1256 |
Instruction Following Qwen2.5 72B Instruct leads
Claude 3 Opus: 64.1 (#228), Qwen2.5 72B Instruct: 65.5 (#221)
| Benchmark | Claude 3 Opus | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Instruction Following | 1248 | 1254 |
| LiveBench Instruction Following | 63.9% | — |
| IFEval | — | 80.6% |
Long Context Too close to call
Claude 3 Opus: 38.2 (#202), Qwen2.5 72B Instruct: 38.9 (#188)
| Benchmark | Claude 3 Opus | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Longer Query | 1259 | 1282 |
Writing & Preference Too close to call
Claude 3 Opus: 47.2 (#213), Qwen2.5 72B Instruct: 46.7 (#215)
| Benchmark | Claude 3 Opus | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Text | 1262 | 1269 |
| LMArena Creative Writing | 1235 | 1221 |
| LMArena Multi-Turn | 1275 | 1272 |
| WildBench | — | 80.2% |
| LiveBench Language | 50.4% | — |
Frequently asked questions
Is Claude 3 Opus better than Qwen2.5 72B Instruct?
Qwen2.5 72B Instruct is the stronger model overall, scoring 31.9 to 29.5 on the Noometry Index.
Is Claude 3 Opus or Qwen2.5 72B Instruct better for coding?
They score almost the same on coding (32.9 vs 33.2); test both on your own repository before choosing.
How many benchmarks do Claude 3 Opus and Qwen2.5 72B Instruct share?
31 benchmarks have published results for both models. Claude 3 Opus has 46 scored results on Noometry and Qwen2.5 72B Instruct has 43.