Model comparison
GPT-5.4 Pro vs Qwen3-30B-A3B
GPT-5.4 Pro is the stronger model overall, scoring 58.9 to 38.9 on the Noometry Index. Qwen3-30B-A3B costs 314× less per token, which makes it the better buy when GPT-5.4 Pro's lead doesn't matter for your workload.
Last verified . 5 shared benchmarks.
Alibaba (Qwen)Summary
- They share 5 benchmarks with published results for both. GPT-5.4 Pro scores higher in 4 categories and Qwen3-30B-A3B in 0 categories; 4 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where GPT-5.4 Pro leads 70.7 to 22.2.
- The biggest single-benchmark swing is Chess Puzzles: 58.6% for GPT-5.4 Pro and 8% for Qwen3-30B-A3B.
- Qwen3-30B-A3B is cheaper at $0.12 / $0.50 per million input/output tokens, against $30 / $180 for GPT-5.4 Pro.
- GPT-5.4 Pro accepts more context: 1.05M tokens versus 41K.
- Qwen3-30B-A3B has downloadable open weights; the other is API-only.
Side by side
| GPT-5.4 Pro | Qwen3-30B-A3B | |
|---|---|---|
| Provider | OpenAI | Alibaba (Qwen) |
| Noometry Index | 58.9 | 38.9 |
| Released | 2026-03-05 | 2025-04-28 |
| Weights | Proprietary | Open |
| Context window | 1.05M | 41K |
| Max output | 128K | 16K |
| Input $ / M tokens | $30 | $0.12 |
| Output $ / M tokens | $180 | $0.50 |
| Results tracked | 16 | 32 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.4 Pro leads
GPT-5.4 Pro: 43.2 (#90), Qwen3-30B-A3B: 37.5 (#194)
| Benchmark | GPT-5.4 Pro | Qwen3-30B-A3B |
|---|---|---|
| WeirdML | 57.4% | 29.8% |
| SciCode | — | 33.3% |
| LMArena Coding | — | 1416 |
Agentic & Tool Use Not comparable
GPT-5.4 Pro: —, Qwen3-30B-A3B: 29.8 (#82)
| Benchmark | GPT-5.4 Pro | Qwen3-30B-A3B |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 41.4% |
Reasoning GPT-5.4 Pro leads
GPT-5.4 Pro: 70.7 (#13), Qwen3-30B-A3B: 22.2 (#204)
| Benchmark | GPT-5.4 Pro | Qwen3-30B-A3B |
|---|---|---|
| CritPt | 30% | 0.3% |
| Chess Puzzles | 58.6% | 8% |
| Epoch Capabilities Index | 158.93 | 139.63 |
| ARC-AGI-2 | 83.3% | — |
| SimpleBench | 74.1% | — |
| Kagi LLM Benchmark | — | 54.9% |
| ARC-AGI-1 | 94.5% | — |
| EnigmaEval | 23.8% | — |
| LMArena Hard Prompts | — | 1398 |
| DTBench | — | 69.3% |
| LMCA | — | 22.4% |
Math GPT-5.4 Pro leads
GPT-5.4 Pro: 72.4 (#22), Qwen3-30B-A3B: 37.4 (#157)
| Benchmark | GPT-5.4 Pro | Qwen3-30B-A3B |
|---|---|---|
| FrontierMath (Tiers 1-3) | 82.5% | — |
| FrontierMath Tier 4 | 58.5% | — |
| MathArena Final-Answer Competitions | — | 47.8% |
| OTIS Mock AIME 2024-2025 | — | 70.3% |
| LMArena Math | — | 1394 |
| FrontierMath (Feb 2025 set) | 50% | — |
| FrontierMath Tier 4 (v1) | 37.5% | — |
Knowledge GPT-5.4 Pro leads
GPT-5.4 Pro: 68.3 (#7), Qwen3-30B-A3B: 41.8 (#105)
| Benchmark | GPT-5.4 Pro | Qwen3-30B-A3B |
|---|---|---|
| GPQA Diamond | 94.6% | 70.1% |
| Humanity's Last Exam | 44.3% | — |
| SimpleQA Verified | 46.3% | — |
| Confabulations | — | 12.3% |
| Vectara Hallucination Rate | 8.3% | — |
| LMArena Expert | — | 1396 |
Multilingual Not comparable
GPT-5.4 Pro: —, Qwen3-30B-A3B: 49.5 (#132)
| Benchmark | GPT-5.4 Pro | Qwen3-30B-A3B |
|---|---|---|
| LMArena Non-English | — | 1372 |
| LMArena Chinese | — | 1433 |
| LMArena French | — | 1418 |
| LMArena German | — | 1380 |
| LMArena Japanese | — | 1337 |
| LMArena Korean | — | 1331 |
| LMArena Russian | — | 1370 |
| LMArena Spanish | — | 1404 |
Instruction Following Not comparable
GPT-5.4 Pro: —, Qwen3-30B-A3B: 72.0 (#142)
| Benchmark | GPT-5.4 Pro | Qwen3-30B-A3B |
|---|---|---|
| LMArena Instruction Following | — | 1363 |
Long Context Not comparable
GPT-5.4 Pro: —, Qwen3-30B-A3B: 31.0 (#283)
| Benchmark | GPT-5.4 Pro | Qwen3-30B-A3B |
|---|---|---|
| Fiction.LiveBench | — | 40.6% |
| LMArena Longer Query | — | 1379 |
Writing & Preference Not comparable
GPT-5.4 Pro: —, Qwen3-30B-A3B: 55.6 (#143)
| Benchmark | GPT-5.4 Pro | Qwen3-30B-A3B |
|---|---|---|
| LMArena Text | — | 1384 |
| LMArena Creative Writing | — | 1317 |
| Short-Story Creative Writing | — | 75.3% |
| LMArena Multi-Turn | — | 1378 |
Frequently asked questions
Is GPT-5.4 Pro better than Qwen3-30B-A3B?
GPT-5.4 Pro is the stronger model overall, scoring 58.9 to 38.9 on the Noometry Index. Qwen3-30B-A3B costs 314× less per token, which makes it the better buy when GPT-5.4 Pro's lead doesn't matter for your workload.
Which is cheaper, GPT-5.4 Pro 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-5.4 Pro lists at $30 and $180.
Is GPT-5.4 Pro or Qwen3-30B-A3B better for coding?
GPT-5.4 Pro scores higher on coding benchmarks: 43.2 versus 37.5 in the Noometry coding category.
Which has the bigger context window?
GPT-5.4 Pro does, with 1.05M tokens against 41K.
How many benchmarks do GPT-5.4 Pro and Qwen3-30B-A3B share?
5 benchmarks have published results for both models. GPT-5.4 Pro has 16 scored results on Noometry and Qwen3-30B-A3B has 32.