Model comparison
GPT-5.4 Pro vs Qwen2.5 72B Instruct
GPT-5.4 Pro is the stronger model overall, scoring 58.9 to 31.9 on the Noometry Index. Qwen2.5 72B Instruct costs 28× less per token, which makes it the better buy when GPT-5.4 Pro's lead doesn't matter for your workload.
Last verified . 3 shared benchmarks.
Alibaba (Qwen)Summary
- They share 3 benchmarks with published results for both. GPT-5.4 Pro scores higher in 4 categories and Qwen2.5 72B Instruct in 0 categories; 4 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-5.4 Pro leads 72.4 to 19.3.
- The biggest single-benchmark swing is GPQA Diamond: 94.6% for GPT-5.4 Pro and 49.1% for Qwen2.5 72B Instruct.
- Qwen2.5 72B Instruct is cheaper at $1.40 / $5.60 per million input/output tokens, against $30 / $180 for GPT-5.4 Pro.
- GPT-5.4 Pro accepts more context: 1.05M tokens versus 131K.
- Qwen2.5 72B Instruct has downloadable open weights; the other is API-only.
Side by side
| GPT-5.4 Pro | Qwen2.5 72B Instruct | |
|---|---|---|
| Provider | OpenAI | Alibaba (Qwen) |
| Noometry Index | 58.9 | 31.9 |
| Released | 2026-03-05 | 2024-09 |
| Weights | Proprietary | Open |
| Context window | 1.05M | 131K |
| Max output | 128K | 8K |
| Input $ / M tokens | $30 | $1.40 |
| Output $ / M tokens | $180 | $5.60 |
| Results tracked | 16 | 43 |
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), Qwen2.5 72B Instruct: 33.2 (#260)
| Benchmark | GPT-5.4 Pro | Qwen2.5 72B Instruct |
|---|---|---|
| WeirdML | 57.4% | 16% |
| BigCodeBench Instruct | — | 45.8% |
| LMArena Coding | — | 1292 |
| BigCodeBench Complete | — | 55.9% |
Agentic & Tool Use Not comparable
GPT-5.4 Pro: —, Qwen2.5 72B Instruct: 22.1 (#133)
| Benchmark | GPT-5.4 Pro | Qwen2.5 72B Instruct |
|---|---|---|
| TheAgentCompany | — | 5.7% |
| BALROG | — | 16.2% |
| METR Time Horizons | — | 35.8% |
Reasoning GPT-5.4 Pro leads
GPT-5.4 Pro: 70.7 (#13), Qwen2.5 72B Instruct: 22.3 (#199)
| Benchmark | GPT-5.4 Pro | Qwen2.5 72B Instruct |
|---|---|---|
| Epoch Capabilities Index | 158.93 | 129 |
| ARC-AGI-2 | 83.3% | — |
| SimpleBench | 74.1% | — |
| ARC-AGI-1 | 94.5% | — |
| CritPt | 30% | — |
| Chess Puzzles | 58.6% | — |
| EnigmaEval | 23.8% | — |
| LMArena Hard Prompts | — | 1271 |
| DTBench | — | 62.9% |
| LMCA | — | 13.4% |
| BIG-Bench Hard | — | 79.8% |
| ForecastBench | — | 57.5 |
| HellaSwag | — | 84.8% |
| PIQA | — | 82.6% |
| WinoGrande | — | 82.3% |
Math GPT-5.4 Pro leads
GPT-5.4 Pro: 72.4 (#22), Qwen2.5 72B Instruct: 19.3 (#287)
| Benchmark | GPT-5.4 Pro | Qwen2.5 72B Instruct |
|---|---|---|
| FrontierMath (Tiers 1-3) | 82.5% | — |
| FrontierMath Tier 4 | 58.5% | — |
| OTIS Mock AIME 2024-2025 | — | 8.1% |
| Omni-MATH | — | 33% |
| LMArena Math | — | 1283 |
| MATH Level 5 | — | 63.2% |
| FrontierMath (Feb 2025 set) | 50% | — |
| FrontierMath Tier 4 (v1) | 37.5% | — |
Knowledge GPT-5.4 Pro leads
GPT-5.4 Pro: 68.3 (#7), Qwen2.5 72B Instruct: 27.0 (#253)
| Benchmark | GPT-5.4 Pro | Qwen2.5 72B Instruct |
|---|---|---|
| GPQA Diamond | 94.6% | 49.1% |
| Humanity's Last Exam | 44.3% | — |
| SimpleQA Verified | 46.3% | — |
| MMLU-Pro | — | 63.1% |
| Confabulations | — | 19.1% |
| Vectara Hallucination Rate | 8.3% | — |
| GPQA (HELM) | — | 42.6% |
| LMArena Expert | — | 1245 |
| ARC (AI2) Challenge | — | 94.5% |
| MMLU | — | 85.3% |
| TriviaQA | — | 71.9% |
Multilingual Not comparable
GPT-5.4 Pro: —, Qwen2.5 72B Instruct: 41.0 (#213)
| Benchmark | GPT-5.4 Pro | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Non-English | — | 1252 |
| LMArena Chinese | — | 1272 |
| LMArena French | — | 1280 |
| LMArena German | — | 1234 |
| LMArena Japanese | — | 1180 |
| LMArena Korean | — | 1188 |
| LMArena Russian | — | 1264 |
| LMArena Spanish | — | 1256 |
Instruction Following Not comparable
GPT-5.4 Pro: —, Qwen2.5 72B Instruct: 65.5 (#221)
| Benchmark | GPT-5.4 Pro | Qwen2.5 72B Instruct |
|---|---|---|
| IFEval | — | 80.6% |
| LMArena Instruction Following | — | 1254 |
Long Context Not comparable
GPT-5.4 Pro: —, Qwen2.5 72B Instruct: 38.9 (#188)
| Benchmark | GPT-5.4 Pro | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Longer Query | — | 1282 |
Writing & Preference Not comparable
GPT-5.4 Pro: —, Qwen2.5 72B Instruct: 46.7 (#215)
| Benchmark | GPT-5.4 Pro | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Text | — | 1269 |
| LMArena Creative Writing | — | 1221 |
| WildBench | — | 80.2% |
| LMArena Multi-Turn | — | 1272 |
Frequently asked questions
Is GPT-5.4 Pro better than Qwen2.5 72B Instruct?
GPT-5.4 Pro is the stronger model overall, scoring 58.9 to 31.9 on the Noometry Index. Qwen2.5 72B Instruct costs 28× 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 Qwen2.5 72B Instruct?
Qwen2.5 72B Instruct is cheaper. It lists at $1.40 per million input tokens and $5.60 per million output tokens; GPT-5.4 Pro lists at $30 and $180.
Is GPT-5.4 Pro or Qwen2.5 72B Instruct better for coding?
GPT-5.4 Pro scores higher on coding benchmarks: 43.2 versus 33.2 in the Noometry coding category.
Which has the bigger context window?
GPT-5.4 Pro does, with 1.05M tokens against 131K.
How many benchmarks do GPT-5.4 Pro and Qwen2.5 72B Instruct share?
3 benchmarks have published results for both models. GPT-5.4 Pro has 16 scored results on Noometry and Qwen2.5 72B Instruct has 43.