Model comparison
GPT-5.4 nano vs Step 3
GPT-5.4 nano is the stronger model overall, scoring 41.9 to 40.5 on the Noometry Index.
Last verified . 17 shared benchmarks.
StepFunSummary
- They share 17 benchmarks with published results for both. GPT-5.4 nano scores higher in 8 categories and Step 3 in 1 category; 9 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where GPT-5.4 nano leads 41.9 to 36.8.
- The biggest single-benchmark swing is Kagi LLM Benchmark: 39.7% for GPT-5.4 nano and 62.3% for Step 3.
- Step 3 has downloadable open weights; the other is API-only.
Side by side
| GPT-5.4 nano | Step 3 | |
|---|---|---|
| Provider | OpenAI | StepFun |
| Noometry Index | 41.9 | 40.5 |
| Released | 2026-03-17 | — |
| Weights | Proprietary | Open |
| Context window | 400K | — |
| Max output | 128K | — |
| Input $ / M tokens | $0.20 | — |
| Output $ / M tokens | $1.25 | — |
| Results tracked | 40 | 17 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.4 nano leads
GPT-5.4 nano: 43.6 (#84), Step 3: 40.1 (#147)
| Benchmark | GPT-5.4 nano | Step 3 |
|---|---|---|
| LMArena Coding | 1405 | 1367 |
| SciCode | 46.9% | — |
| WeirdML | 49.2% | — |
| ALE-Bench | 1,005 | — |
Reasoning Step 3 leads
GPT-5.4 nano: 23.7 (#173), Step 3: 28.4 (#105)
| Benchmark | GPT-5.4 nano | Step 3 |
|---|---|---|
| Kagi LLM Benchmark | 39.7% | 62.3% |
| LMArena Hard Prompts | 1381 | 1355 |
| ARC-AGI-2 | 5.7% | — |
| ARC-AGI-1 | 51.5% | — |
| CritPt | 9.3% | — |
| Chess Puzzles | 30% | — |
| Mystery Game Puzzles | 9% | — |
| DTBench | 80.3% | — |
| LMCA | 36.9% | — |
| Epoch Capabilities Index | 145.81 | — |
| ForecastBench | 57.3 | — |
Math GPT-5.4 nano leads
GPT-5.4 nano: 40.9 (#88), Step 3: 37.6 (#148)
| Benchmark | GPT-5.4 nano | Step 3 |
|---|---|---|
| LMArena Math | 1406 | 1366 |
| FrontierMath (Tiers 1-3) | 44.9% | — |
| FrontierMath Tier 4 | 12.2% | — |
| OTIS Mock AIME 2024-2025 | 87.8% | — |
| ProofBench | 5% | — |
| FrontierMath (Feb 2025 set) | 25.9% | — |
| FrontierMath Tier 4 (v1) | 6.3% | — |
Knowledge GPT-5.4 nano leads
GPT-5.4 nano: 41.9 (#103), Step 3: 36.8 (#164)
| Benchmark | GPT-5.4 nano | Step 3 |
|---|---|---|
| LMArena Expert | 1396 | 1333 |
| GPQA Diamond | 78.5% | — |
| SimpleQA Verified | 11.7% | — |
| Vectara Hallucination Rate | 3.1% | — |
Multimodal GPT-5.4 nano leads
GPT-5.4 nano: 36.7 (#78), Step 3: 35.5 (#86)
| Benchmark | GPT-5.4 nano | Step 3 |
|---|---|---|
| LMArena Vision | 1196 | 1177 |
Multilingual GPT-5.4 nano leads
GPT-5.4 nano: 48.6 (#140), Step 3: 46.3 (#159)
| Benchmark | GPT-5.4 nano | Step 3 |
|---|---|---|
| LMArena Non-English | 1359 | 1327 |
| LMArena Chinese | 1392 | 1397 |
| LMArena German | 1367 | 1371 |
| LMArena Korean | 1320 | 1269 |
| LMArena Russian | 1363 | 1331 |
| LMArena Spanish | 1371 | 1371 |
| LMArena French | 1396 | — |
| LMArena Japanese | 1343 | — |
Instruction Following GPT-5.4 nano leads
GPT-5.4 nano: 71.9 (#144), Step 3: 70.4 (#164)
| Benchmark | GPT-5.4 nano | Step 3 |
|---|---|---|
| LMArena Instruction Following | 1362 | 1332 |
Long Context GPT-5.4 nano leads
GPT-5.4 nano: 41.6 (#137), Step 3: 40.3 (#157)
| Benchmark | GPT-5.4 nano | Step 3 |
|---|---|---|
| LMArena Longer Query | 1366 | 1326 |
Writing & Preference GPT-5.4 nano leads
GPT-5.4 nano: 55.7 (#142), Step 3: 54.3 (#151)
| Benchmark | GPT-5.4 nano | Step 3 |
|---|---|---|
| LMArena Text | 1372 | 1350 |
| LMArena Creative Writing | 1314 | 1321 |
| LMArena Multi-Turn | 1382 | 1341 |
Frequently asked questions
Is GPT-5.4 nano better than Step 3?
GPT-5.4 nano is the stronger model overall, scoring 41.9 to 40.5 on the Noometry Index.
Is GPT-5.4 nano or Step 3 better for coding?
GPT-5.4 nano scores higher on coding benchmarks: 43.6 versus 40.1 in the Noometry coding category.
How many benchmarks do GPT-5.4 nano and Step 3 share?
17 benchmarks have published results for both models. GPT-5.4 nano has 40 scored results on Noometry and Step 3 has 17.