Model comparison
GPT-5 Pro vs Step 3
GPT-5 Pro is the stronger model overall, scoring 46.4 to 40.5 on the Noometry Index.
Last verified . 1 shared benchmarks.
StepFunSummary
- They share 1 benchmark with published results for both. GPT-5 Pro scores higher in 4 categories and Step 3 in 0 categories; 4 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where GPT-5 Pro leads 56.7 to 36.8.
- The biggest single-benchmark swing is Kagi LLM Benchmark: 76.8% for GPT-5 Pro and 62.3% for Step 3.
- Step 3 has downloadable open weights; the other is API-only.
Side by side
| GPT-5 Pro | Step 3 | |
|---|---|---|
| Provider | OpenAI | StepFun |
| Noometry Index | 46.4 | 40.5 |
| Released | 2025-10-06 | — |
| Weights | Proprietary | Open |
| Context window | 400K | — |
| Max output | 272K | — |
| Input $ / M tokens | $15 | — |
| Output $ / M tokens | $120 | — |
| Results tracked | 12 | 17 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5 Pro leads
GPT-5 Pro: 44.0 (#80), Step 3: 40.1 (#147)
| Benchmark | GPT-5 Pro | Step 3 |
|---|---|---|
| WeirdML | 60.4% | — |
| LMArena Coding | — | 1367 |
| AlgoTune | 1.31 | — |
Reasoning GPT-5 Pro leads
GPT-5 Pro: 38.9 (#62), Step 3: 28.4 (#105)
| Benchmark | GPT-5 Pro | Step 3 |
|---|---|---|
| Kagi LLM Benchmark | 76.8% | 62.3% |
| ARC-AGI-2 | 18.3% | — |
| SimpleBench | 61.6% | — |
| ARC-AGI-1 | 70.2% | — |
| EnigmaEval | 18.8% | — |
| LMArena Hard Prompts | — | 1355 |
| Epoch Capabilities Index | 150.28 | — |
Math GPT-5 Pro leads
GPT-5 Pro: 48.5 (#63), Step 3: 37.6 (#148)
| Benchmark | GPT-5 Pro | Step 3 |
|---|---|---|
| FrontierMath (Tiers 1-3) | 55.8% | — |
| FrontierMath Tier 4 | 19.5% | — |
| LMArena Math | — | 1366 |
| FrontierMath Tier 4 (v1) | 14.6% | — |
Knowledge GPT-5 Pro leads
GPT-5 Pro: 56.7 (#42), Step 3: 36.8 (#164)
| Benchmark | GPT-5 Pro | Step 3 |
|---|---|---|
| Humanity's Last Exam | 31.6% | — |
| LMArena Expert | — | 1333 |
Multimodal Not comparable
GPT-5 Pro: —, Step 3: 35.5 (#86)
| Benchmark | GPT-5 Pro | Step 3 |
|---|---|---|
| LMArena Vision | — | 1177 |
Multilingual Not comparable
GPT-5 Pro: —, Step 3: 46.3 (#159)
| Benchmark | GPT-5 Pro | Step 3 |
|---|---|---|
| LMArena Non-English | — | 1327 |
| LMArena Chinese | — | 1397 |
| LMArena German | — | 1371 |
| LMArena Korean | — | 1269 |
| LMArena Russian | — | 1331 |
| LMArena Spanish | — | 1371 |
Instruction Following Not comparable
GPT-5 Pro: —, Step 3: 70.4 (#164)
| Benchmark | GPT-5 Pro | Step 3 |
|---|---|---|
| LMArena Instruction Following | — | 1332 |
Long Context Not comparable
GPT-5 Pro: —, Step 3: 40.3 (#157)
| Benchmark | GPT-5 Pro | Step 3 |
|---|---|---|
| LMArena Longer Query | — | 1326 |
Writing & Preference Not comparable
GPT-5 Pro: —, Step 3: 54.3 (#151)
| Benchmark | GPT-5 Pro | Step 3 |
|---|---|---|
| LMArena Text | — | 1350 |
| LMArena Creative Writing | — | 1321 |
| LMArena Multi-Turn | — | 1341 |
Frequently asked questions
Is GPT-5 Pro better than Step 3?
GPT-5 Pro is the stronger model overall, scoring 46.4 to 40.5 on the Noometry Index.
Is GPT-5 Pro or Step 3 better for coding?
GPT-5 Pro scores higher on coding benchmarks: 44.0 versus 40.1 in the Noometry coding category.
How many benchmarks do GPT-5 Pro and Step 3 share?
1 benchmark has published results for both models. GPT-5 Pro has 12 scored results on Noometry and Step 3 has 17.