Model comparison
GPT-4.1 vs Qwen3-30B-A3B
Qwen3-30B-A3B is the stronger model overall, scoring 38.5 to 35.7 on the Noometry Index.
Last verified . 27 shared benchmarks.
Alibaba (Qwen)Summary
- They share 27 benchmarks with published results for both. GPT-4.1 scores higher in 3 categories and Qwen3-30B-A3B in 6 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in math, where Qwen3-30B-A3B leads 36.0 to 21.4.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 38.3% for GPT-4.1 and 70.3% for Qwen3-30B-A3B.
- Qwen3-30B-A3B is cheaper at $0.12 / $0.50 per million input/output tokens, against $2 / $8 for GPT-4.1.
- GPT-4.1 accepts more context: 1.05M tokens versus 41K.
- Qwen3-30B-A3B has downloadable open weights; the other is API-only.
Side by side
| GPT-4.1 | Qwen3-30B-A3B | |
|---|---|---|
| Provider | OpenAI | Alibaba (Qwen) |
| Noometry Index | 35.7 | 38.5 |
| Released | 2025-04-14 | 2025-04-28 |
| Weights | Proprietary | Open |
| Context window | 1.05M | 41K |
| Max output | 33K | 16K |
| Input $ / M tokens | $2 | $0.12 |
| Output $ / M tokens | $8 | $0.50 |
| Results tracked | 52 | 32 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3-30B-A3B leads
GPT-4.1: 34.2 (#228), Qwen3-30B-A3B: 37.0 (#188)
| Benchmark | GPT-4.1 | Qwen3-30B-A3B |
|---|---|---|
| WeirdML | 39% | 29.8% |
| LMArena Coding | 1391 | 1416 |
| SWE-bench Verified | 48.5% | — |
| SWE-bench Verified (bash only) | 39.6% | — |
| Aider Polyglot | 52.4% | — |
| SciCode | — | 33.3% |
| CadEval | 42% | — |
| ALE-Bench | 558.1 | — |
Agentic & Tool Use GPT-4.1 leads
GPT-4.1: 34.4 (#40), Qwen3-30B-A3B: 29.5 (#75)
| Benchmark | GPT-4.1 | Qwen3-30B-A3B |
|---|---|---|
| Berkeley Function Calling Leaderboard | 54% | 41.4% |
Reasoning Qwen3-30B-A3B leads
GPT-4.1: 11.7 (#334), Qwen3-30B-A3B: 21.6 (#193)
| Benchmark | GPT-4.1 | Qwen3-30B-A3B |
|---|---|---|
| Kagi LLM Benchmark | 52.3% | 54.9% |
| Chess Puzzles | 6% | 8% |
| LMArena Hard Prompts | 1384 | 1398 |
| DTBench | 68.3% | 69.3% |
| LMCA | 25.6% | 22.4% |
| Epoch Capabilities Index | 136.79 | 139.63 |
| ARC-AGI-2 | 0.4% | — |
| SimpleBench | 27% | — |
| ARC-AGI-1 | 5.5% | — |
| CritPt | — | 0.3% |
| EnigmaEval | 2.2% | — |
| ForecastBench | 61.5 | — |
Math Qwen3-30B-A3B leads
GPT-4.1: 21.4 (#269), Qwen3-30B-A3B: 36.0 (#133)
| Benchmark | GPT-4.1 | Qwen3-30B-A3B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 38.3% | 70.3% |
| LMArena Math | 1369 | 1394 |
| FrontierMath (Tiers 1-3) | 6% | — |
| MathArena Final-Answer Competitions | — | 47.8% |
| Omni-MATH | 47.1% | — |
| MATH Level 5 | 83% | — |
| FrontierMath (Feb 2025 set) | 5.5% | — |
| FrontierMath Tier 4 (v1) | 0% | — |
Knowledge Qwen3-30B-A3B leads
GPT-4.1: 37.0 (#152), Qwen3-30B-A3B: 41.7 (#96)
| Benchmark | GPT-4.1 | Qwen3-30B-A3B |
|---|---|---|
| GPQA Diamond | 66.9% | 70.1% |
| LMArena Expert | 1364 | 1395 |
| Humanity's Last Exam | 5.4% | — |
| SimpleQA Verified | 31.1% | — |
| MMLU-Pro | 81.1% | — |
| Confabulations | — | 12.3% |
| Vectara Hallucination Rate | 5.6% | — |
| GPQA (HELM) | 65.9% | — |
Multimodal Not comparable
GPT-4.1: 37.5 (#66), Qwen3-30B-A3B: —
| Benchmark | GPT-4.1 | Qwen3-30B-A3B |
|---|---|---|
| LMArena Vision | 1210 | — |
| GeoBench | 72% | — |
Multilingual Too close to call
GPT-4.1: 49.6 (#128), Qwen3-30B-A3B: 49.8 (#126)
| Benchmark | GPT-4.1 | Qwen3-30B-A3B |
|---|---|---|
| LMArena Non-English | 1370 | 1372 |
| LMArena Chinese | 1382 | 1434 |
| LMArena French | 1383 | 1419 |
| LMArena German | 1380 | 1379 |
| LMArena Japanese | 1320 | 1338 |
| LMArena Korean | 1339 | 1333 |
| LMArena Russian | 1377 | 1369 |
| LMArena Spanish | 1375 | 1403 |
Instruction Following Too close to call
GPT-4.1: 71.5 (#148), Qwen3-30B-A3B: 72.2 (#137)
| Benchmark | GPT-4.1 | Qwen3-30B-A3B |
|---|---|---|
| LMArena Instruction Following | 1367 | 1364 |
| IFEval | 83.8% | — |
Long Context GPT-4.1 leads
GPT-4.1: 40.0 (#158), Qwen3-30B-A3B: 31.0 (#278)
| Benchmark | GPT-4.1 | Qwen3-30B-A3B |
|---|---|---|
| Fiction.LiveBench | 63.9% | 40.6% |
| LMArena Longer Query | 1385 | 1379 |
Writing & Preference GPT-4.1 leads
GPT-4.1: 57.8 (#120), Qwen3-30B-A3B: 55.9 (#138)
| Benchmark | GPT-4.1 | Qwen3-30B-A3B |
|---|---|---|
| LMArena Text | 1383 | 1384 |
| LMArena Creative Writing | 1363 | 1318 |
| LMArena Multi-Turn | 1398 | 1378 |
| Short-Story Creative Writing | — | 75.3% |
| EQ-Bench Creative Writing | 1420 | — |
| WildBench | 85.4% | — |
Frequently asked questions
Is GPT-4.1 better than Qwen3-30B-A3B?
Qwen3-30B-A3B is the stronger model overall, scoring 38.5 to 35.7 on the Noometry Index.
Which is cheaper, GPT-4.1 or Qwen3-30B-A3B?
Qwen3-30B-A3B is cheaper. It lists at $0.12 per million input tokens and $0.50 per million output tokens; GPT-4.1 lists at $2 and $8.
Is GPT-4.1 or Qwen3-30B-A3B better for coding?
Qwen3-30B-A3B scores higher on coding benchmarks: 37.0 versus 34.2 in the Noometry coding category.
Which has the bigger context window?
GPT-4.1 does, with 1.05M tokens against 41K.
How many benchmarks do GPT-4.1 and Qwen3-30B-A3B share?
27 benchmarks have published results for both models. GPT-4.1 has 52 scored results on Noometry and Qwen3-30B-A3B has 32.