Model comparison
GPT-5.5 Instant vs Step 5 Preview
Step 5 Preview is the stronger model overall, scoring 47.9 to 42.7 on the Noometry Index.
Last verified . 16 shared benchmarks.
StepFunSummary
- They share 16 benchmarks with published results for both. GPT-5.5 Instant scores higher in 1 category and Step 5 Preview in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Step 5 Preview leads 46.1 to 26.5.
- The biggest single-benchmark swing is CritPt: 0% for GPT-5.5 Instant and 20.9% for Step 5 Preview.
Side by side
| GPT-5.5 Instant | Step 5 Preview | |
|---|---|---|
| Provider | OpenAI | StepFun |
| Noometry Index | 42.7 | 47.9 |
| Released | 2026-05-05 | 2026-09-16 |
| Weights | Proprietary | Proprietary |
| Context window | — | 1.02M |
| Max output | — | 66K |
| Input $ / M tokens | — | $1 |
| Output $ / M tokens | — | $2.70 |
| Results tracked | 27 | 18 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Step 5 Preview leads
GPT-5.5 Instant: 44.3 (#74), Step 5 Preview: 51.8 (#36)
| Benchmark | GPT-5.5 Instant | Step 5 Preview |
|---|---|---|
| SciCode | 48.6% | 58.9% |
| LMArena Coding | 1433 | 1480 |
| LMArena WebDev | — | 1564 |
Reasoning Step 5 Preview leads
GPT-5.5 Instant: 24.9 (#155), Step 5 Preview: 40.0 (#57)
| Benchmark | GPT-5.5 Instant | Step 5 Preview |
|---|---|---|
| CritPt | 0% | 20.9% |
| LMArena Hard Prompts | 1426 | 1465 |
| Chess Puzzles | 12% | — |
| Epoch Capabilities Index | 142.52 | — |
Math Step 5 Preview leads
GPT-5.5 Instant: 26.5 (#259), Step 5 Preview: 46.1 (#72)
| Benchmark | GPT-5.5 Instant | Step 5 Preview |
|---|---|---|
| LMArena Math | 1420 | 1470 |
| FrontierMath (Tiers 1-3) | 26.3% | — |
| FrontierMath Tier 4 | 2.4% | — |
| OTIS Mock AIME 2024-2025 | 68.1% | — |
| ProofBench | — | 42% |
Knowledge GPT-5.5 Instant leads
GPT-5.5 Instant: 48.9 (#74), Step 5 Preview: 41.2 (#112)
| Benchmark | GPT-5.5 Instant | Step 5 Preview |
|---|---|---|
| LMArena Expert | 1409 | 1470 |
| GPQA Diamond | 82.5% | — |
Multimodal Step 5 Preview leads
GPT-5.5 Instant: 40.0 (#52), Step 5 Preview: 41.0 (#41)
| Benchmark | GPT-5.5 Instant | Step 5 Preview |
|---|---|---|
| LMArena Vision | 1250 | 1267 |
| LMArena Document | 1403 | — |
Multilingual Step 5 Preview leads
GPT-5.5 Instant: 52.8 (#80), Step 5 Preview: 53.8 (#54)
| Benchmark | GPT-5.5 Instant | Step 5 Preview |
|---|---|---|
| LMArena Non-English | 1417 | 1432 |
| LMArena Chinese | 1456 | 1519 |
| LMArena Russian | 1431 | 1436 |
| LMArena Spanish | 1429 | 1439 |
| LMArena French | 1428 | — |
| LMArena German | 1411 | — |
| LMArena Japanese | 1408 | — |
| LMArena Korean | 1392 | — |
Instruction Following Step 5 Preview leads
GPT-5.5 Instant: 74.2 (#100), Step 5 Preview: 76.0 (#50)
| Benchmark | GPT-5.5 Instant | Step 5 Preview |
|---|---|---|
| LMArena Instruction Following | 1406 | 1444 |
Long Context Step 5 Preview leads
GPT-5.5 Instant: 43.4 (#96), Step 5 Preview: 44.5 (#64)
| Benchmark | GPT-5.5 Instant | Step 5 Preview |
|---|---|---|
| LMArena Longer Query | 1422 | 1455 |
Writing & Preference Too close to call
GPT-5.5 Instant: 61.8 (#85), Step 5 Preview: 62.8 (#71)
| Benchmark | GPT-5.5 Instant | Step 5 Preview |
|---|---|---|
| LMArena Text | 1419 | 1442 |
| LMArena Creative Writing | 1419 | 1410 |
| LMArena Multi-Turn | 1433 | 1447 |
Frequently asked questions
Is GPT-5.5 Instant better than Step 5 Preview?
Step 5 Preview is the stronger model overall, scoring 47.9 to 42.7 on the Noometry Index.
Is GPT-5.5 Instant or Step 5 Preview better for coding?
Step 5 Preview scores higher on coding benchmarks: 51.8 versus 44.3 in the Noometry coding category.
How many benchmarks do GPT-5.5 Instant and Step 5 Preview share?
16 benchmarks have published results for both models. GPT-5.5 Instant has 27 scored results on Noometry and Step 5 Preview has 18.