Model comparison
GPT-4.5 vs Qwen3-30B-A3B
Qwen3-30B-A3B is the stronger model overall, scoring 38.5 to 36.9 on the Noometry Index.
Last verified . 23 shared benchmarks.
Alibaba (Qwen)Summary
- They share 23 benchmarks with published results for both. GPT-4.5 scores higher in 5 categories and Qwen3-30B-A3B in 4 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in long context, where GPT-4.5 leads 40.4 to 31.0.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 37.8% for GPT-4.5 and 70.3% for Qwen3-30B-A3B.
- Qwen3-30B-A3B has downloadable open weights; the other is API-only.
Side by side
| GPT-4.5 | Qwen3-30B-A3B | |
|---|---|---|
| Provider | OpenAI | Alibaba (Qwen) |
| Noometry Index | 36.9 | 38.5 |
| Released | 2025-02-27 | 2025-04-28 |
| Weights | Proprietary | Open |
| Context window | — | 41K |
| Max output | — | 16K |
| Input $ / M tokens | — | $0.12 |
| Output $ / M tokens | — | $0.50 |
| Results tracked | 42 | 32 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-4.5 leads
GPT-4.5: 41.8 (#102), Qwen3-30B-A3B: 37.0 (#188)
| Benchmark | GPT-4.5 | Qwen3-30B-A3B |
|---|---|---|
| WeirdML | 39.4% | 29.8% |
| LMArena Coding | 1397 | 1416 |
| Aider Polyglot | 44.9% | — |
| SciCode | — | 33.3% |
| LiveBench Coding | 75.2% | — |
Agentic & Tool Use Qwen3-30B-A3B leads
GPT-4.5: 27.6 (#92), Qwen3-30B-A3B: 29.5 (#75)
| Benchmark | GPT-4.5 | Qwen3-30B-A3B |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 41.4% |
| Cybench | 17.5% | — |
Reasoning Qwen3-30B-A3B leads
GPT-4.5: 14.1 (#320), Qwen3-30B-A3B: 21.6 (#193)
| Benchmark | GPT-4.5 | Qwen3-30B-A3B |
|---|---|---|
| LMArena Hard Prompts | 1404 | 1398 |
| Epoch Capabilities Index | 136.74 | 139.63 |
| ARC-AGI-2 | 0.8% | — |
| SimpleBench | 34.5% | — |
| Kagi LLM Benchmark | — | 54.9% |
| ARC-AGI-1 | 10.3% | — |
| CritPt | — | 0.3% |
| Chess Puzzles | — | 8% |
| EnigmaEval | 3.2% | — |
| LiveBench Reasoning | 71.1% | — |
| DTBench | — | 69.3% |
| LiveBench Data Analysis | 64.3% | — |
| LMCA | — | 22.4% |
| ForecastBench | 61.7 | — |
| LiveBench | 69% | — |
Math Qwen3-30B-A3B leads
GPT-4.5: 31.0 (#200), Qwen3-30B-A3B: 36.0 (#133)
| Benchmark | GPT-4.5 | Qwen3-30B-A3B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 37.8% | 70.3% |
| LMArena Math | 1412 | 1394 |
| MathArena Final-Answer Competitions | — | 47.8% |
| LiveBench Math | 69.3% | — |
| MATH Level 5 | 78.6% | — |
Knowledge Qwen3-30B-A3B leads
GPT-4.5: 32.4 (#205), Qwen3-30B-A3B: 41.7 (#96)
| Benchmark | GPT-4.5 | Qwen3-30B-A3B |
|---|---|---|
| GPQA Diamond | 68.7% | 70.1% |
| Confabulations | 13.6% | 12.3% |
| LMArena Expert | 1394 | 1395 |
| Humanity's Last Exam | 5.4% | — |
Multimodal Not comparable
GPT-4.5: 37.0 (#68), Qwen3-30B-A3B: —
| Benchmark | GPT-4.5 | Qwen3-30B-A3B |
|---|---|---|
| LMArena Vision | 1195 | — |
| VPCT | 45% | — |
Multilingual GPT-4.5 leads
GPT-4.5: 52.7 (#78), Qwen3-30B-A3B: 49.8 (#126)
| Benchmark | GPT-4.5 | Qwen3-30B-A3B |
|---|---|---|
| LMArena Non-English | 1413 | 1372 |
| LMArena Chinese | 1421 | 1434 |
| LMArena French | 1417 | 1419 |
| LMArena German | 1457 | 1379 |
| LMArena Japanese | 1416 | 1338 |
| LMArena Korean | 1392 | 1333 |
| LMArena Russian | 1419 | 1369 |
| LMArena Spanish | — | 1403 |
Instruction Following Too close to call
GPT-4.5: 72.8 (#129), Qwen3-30B-A3B: 72.2 (#137)
| Benchmark | GPT-4.5 | Qwen3-30B-A3B |
|---|---|---|
| LMArena Instruction Following | 1404 | 1364 |
| LiveBench Instruction Following | 72.3% | — |
Long Context GPT-4.5 leads
GPT-4.5: 40.4 (#150), Qwen3-30B-A3B: 31.0 (#278)
| Benchmark | GPT-4.5 | Qwen3-30B-A3B |
|---|---|---|
| Fiction.LiveBench | 63.9% | 40.6% |
| LMArena Longer Query | 1406 | 1379 |
Writing & Preference GPT-4.5 leads
GPT-4.5: 57.1 (#129), Qwen3-30B-A3B: 55.9 (#138)
| Benchmark | GPT-4.5 | Qwen3-30B-A3B |
|---|---|---|
| LMArena Text | 1417 | 1384 |
| LMArena Creative Writing | 1394 | 1318 |
| Short-Story Creative Writing | 75.6% | 75.3% |
| LMArena Multi-Turn | 1444 | 1378 |
| EQ-Bench Creative Writing | 1258 | — |
| LiveBench Language | 61.5% | — |
Frequently asked questions
Is GPT-4.5 better than Qwen3-30B-A3B?
Qwen3-30B-A3B is the stronger model overall, scoring 38.5 to 36.9 on the Noometry Index.
Is GPT-4.5 or Qwen3-30B-A3B better for coding?
GPT-4.5 scores higher on coding benchmarks: 41.8 versus 37.0 in the Noometry coding category.
How many benchmarks do GPT-4.5 and Qwen3-30B-A3B share?
23 benchmarks have published results for both models. GPT-4.5 has 42 scored results on Noometry and Qwen3-30B-A3B has 32.