Model comparison
Gemini 3 Pro vs Qwen2.5 72B Instruct
Gemini 3 Pro is the stronger model overall, scoring 54.8 to 31.9 on the Noometry Index.
Last verified . 29 shared benchmarks.
Google
Alibaba (Qwen)Summary
- They share 29 benchmarks with published results for both. Gemini 3 Pro scores higher in 9 categories and Qwen2.5 72B Instruct in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Gemini 3 Pro leads 64.4 to 27.0.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 91.4% for Gemini 3 Pro and 8.1% for Qwen2.5 72B Instruct.
- Qwen2.5 72B Instruct has downloadable open weights; the other is API-only.
Side by side
| Gemini 3 Pro | Qwen2.5 72B Instruct | |
|---|---|---|
| Provider | Alibaba (Qwen) | |
| Noometry Index | 54.8 | 31.9 |
| Released | 2025-11-18 | 2024-09 |
| Weights | Proprietary | Open |
| Context window | — | 131K |
| Max output | — | 8K |
| Input $ / M tokens | — | $1.40 |
| Output $ / M tokens | — | $5.60 |
| Results tracked | 67 | 43 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Gemini 3 Pro leads
Gemini 3 Pro: 51.6 (#39), Qwen2.5 72B Instruct: 33.2 (#260)
| Benchmark | Gemini 3 Pro | Qwen2.5 72B Instruct |
|---|---|---|
| WeirdML | 69.9% | 16% |
| LMArena Coding | 1481 | 1292 |
| SWE-bench Verified | 72.9% | — |
| SWE-bench Verified (bash only) | 74.2% | — |
| LMArena WebDev | 1440 | — |
| SWE-bench Multilingual | 68.7% | — |
| GSO | 18.6% | — |
| BigCodeBench Instruct | — | 45.8% |
| BigCodeBench Complete | — | 55.9% |
| ALE-Bench | 1,177 | — |
| AlgoTune | 1.83 | — |
Agentic & Tool Use Gemini 3 Pro leads
Gemini 3 Pro: 40.6 (#23), Qwen2.5 72B Instruct: 22.1 (#133)
| Benchmark | Gemini 3 Pro | Qwen2.5 72B Instruct |
|---|---|---|
| BALROG | 58.1% | 16.2% |
| METR Time Horizons | 71% | 35.8% |
| Terminal-Bench | 69.4% | — |
| Berkeley Function Calling Leaderboard | 72.5% | — |
| GDPval | 40.3% | — |
| Remote Labor Index | 1.3% | — |
| TheAgentCompany | — | 5.7% |
| τ²-bench Airline | 80.5% | — |
| τ²-bench Banking | 18% | — |
| τ²-bench Retail | 75.9% | — |
| τ²-bench Telecom | 91% | — |
| DeepResearch Bench | 46.3% | — |
| LMArena Search | 1207 | — |
| Vending-Bench 2 | 5,478 | — |
Reasoning Gemini 3 Pro leads
Gemini 3 Pro: 52.5 (#31), Qwen2.5 72B Instruct: 22.3 (#199)
| Benchmark | Gemini 3 Pro | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Hard Prompts | 1480 | 1271 |
| Epoch Capabilities Index | 152.92 | 129 |
| ForecastBench | 61.2 | 57.5 |
| ARC-AGI-2 | 31.1% | — |
| SimpleBench | 76.4% | — |
| Kagi LLM Benchmark | 80.1% | — |
| NYT Connections (extended) | 94.4% | — |
| ARC-AGI-1 | 75% | — |
| CritPt | 6.9% | — |
| Chess Puzzles | 31% | — |
| EnigmaEval | 18.2% | — |
| DTBench | — | 62.9% |
| LMCA | — | 13.4% |
| BIG-Bench Hard | — | 79.8% |
| HellaSwag | — | 84.8% |
| PIQA | — | 82.6% |
| WinoGrande | — | 82.3% |
Math Gemini 3 Pro leads
Gemini 3 Pro: 49.9 (#59), Qwen2.5 72B Instruct: 19.3 (#287)
| Benchmark | Gemini 3 Pro | Qwen2.5 72B Instruct |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 91.4% | 8.1% |
| Omni-MATH | 55.5% | 33% |
| LMArena Math | 1476 | 1283 |
| MathArena Final-Answer Competitions | 67% | — |
| ProofBench | 20% | — |
| MATH Level 5 | — | 63.2% |
| FrontierMath (Feb 2025 set) | 37.6% | — |
| FrontierMath Tier 4 (v1) | 18.8% | — |
Knowledge Gemini 3 Pro leads
Gemini 3 Pro: 64.4 (#16), Qwen2.5 72B Instruct: 27.0 (#253)
| Benchmark | Gemini 3 Pro | Qwen2.5 72B Instruct |
|---|---|---|
| GPQA Diamond | 92.6% | 49.1% |
| MMLU-Pro | 90.3% | 63.1% |
| GPQA (HELM) | 80.3% | 42.6% |
| LMArena Expert | 1475 | 1245 |
| Humanity's Last Exam | 37.5% | — |
| Confabulations | — | 19.1% |
| Vectara Hallucination Rate | 13.6% | — |
| ARC (AI2) Challenge | — | 94.5% |
| MMLU | — | 85.3% |
| TriviaQA | — | 71.9% |
Multimodal Not comparable
Gemini 3 Pro: 57.6 (#2), Qwen2.5 72B Instruct: —
| Benchmark | Gemini 3 Pro | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Vision | 1305 | — |
| GeoBench | 84% | — |
| VPCT | 91% | — |
| LMArena Document | 1434 | — |
Multilingual Gemini 3 Pro leads
Gemini 3 Pro: 56.9 (#16), Qwen2.5 72B Instruct: 41.0 (#213)
| Benchmark | Gemini 3 Pro | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Non-English | 1474 | 1252 |
| LMArena Chinese | 1523 | 1272 |
| LMArena French | 1492 | 1280 |
| LMArena German | 1515 | 1234 |
| LMArena Japanese | 1510 | 1180 |
| LMArena Korean | 1448 | 1188 |
| LMArena Russian | 1493 | 1264 |
| LMArena Spanish | 1470 | 1256 |
Instruction Following Gemini 3 Pro leads
Gemini 3 Pro: 76.3 (#45), Qwen2.5 72B Instruct: 65.5 (#221)
| Benchmark | Gemini 3 Pro | Qwen2.5 72B Instruct |
|---|---|---|
| IFEval | 87.7% | 80.6% |
| LMArena Instruction Following | 1458 | 1254 |
Long Context Gemini 3 Pro leads
Gemini 3 Pro: 44.0 (#79), Qwen2.5 72B Instruct: 38.9 (#188)
| Benchmark | Gemini 3 Pro | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Longer Query | 1471 | 1282 |
| CL-bench | 15.8% | — |
Writing & Preference Gemini 3 Pro leads
Gemini 3 Pro: 66.4 (#35), Qwen2.5 72B Instruct: 46.7 (#215)
| Benchmark | Gemini 3 Pro | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Text | 1479 | 1269 |
| LMArena Creative Writing | 1482 | 1221 |
| WildBench | 85.9% | 80.2% |
| LMArena Multi-Turn | 1484 | 1272 |
| EQ-Bench Creative Writing | 1525 | — |
Frequently asked questions
Is Gemini 3 Pro better than Qwen2.5 72B Instruct?
Gemini 3 Pro is the stronger model overall, scoring 54.8 to 31.9 on the Noometry Index.
Is Gemini 3 Pro or Qwen2.5 72B Instruct better for coding?
Gemini 3 Pro scores higher on coding benchmarks: 51.6 versus 33.2 in the Noometry coding category.
How many benchmarks do Gemini 3 Pro and Qwen2.5 72B Instruct share?
29 benchmarks have published results for both models. Gemini 3 Pro has 67 scored results on Noometry and Qwen2.5 72B Instruct has 43.