Model comparison
GPT-5.3 Codex vs Step 5 Preview
Step 5 Preview is the stronger model overall, scoring 47.9 to 45.8 on the Noometry Index.
Last verified . 1 shared benchmarks.
StepFunSummary
- They share 1 benchmark with published results for both. GPT-5.3 Codex scores higher in 0 categories and Step 5 Preview in 1 category; one gap is clear of the uncertainty.
- The widest gap is in coding, where Step 5 Preview leads 51.8 to 48.6.
- Step 5 Preview is cheaper at $1 / $2.70 per million input/output tokens, against $1.75 / $14 for GPT-5.3 Codex.
- Step 5 Preview accepts more context: 1.02M tokens versus 400K.
Side by side
| GPT-5.3 Codex | Step 5 Preview | |
|---|---|---|
| Provider | OpenAI | StepFun |
| Noometry Index | 45.8 | 47.9 |
| Released | 2026-02-05 | 2026-09-16 |
| Weights | Proprietary | Proprietary |
| Context window | 400K | 1.02M |
| Max output | 128K | 66K |
| Input $ / M tokens | $1.75 | $1 |
| Output $ / M tokens | $14 | $2.70 |
| Results tracked | 8 | 18 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Step 5 Preview leads
GPT-5.3 Codex: 48.6 (#56), Step 5 Preview: 51.8 (#36)
| Benchmark | GPT-5.3 Codex | Step 5 Preview |
|---|---|---|
| LMArena WebDev | 1409 | 1564 |
| SWE-bench Verified | 74.8% | — |
| SciCode | — | 58.9% |
| WeirdML | 79.3% | — |
| LMArena Coding | — | 1480 |
| ALE-Bench | 1,655 | — |
Agentic & Tool Use Not comparable
GPT-5.3 Codex: 48.0 (#9), Step 5 Preview: —
| Benchmark | GPT-5.3 Codex | Step 5 Preview |
|---|---|---|
| Terminal-Bench | 78.4% | — |
| METR Time Horizons | 74.5% | — |
| Vending-Bench 2 | 5,940 | — |
Reasoning Not comparable
GPT-5.3 Codex: —, Step 5 Preview: 40.0 (#57)
| Benchmark | GPT-5.3 Codex | Step 5 Preview |
|---|---|---|
| CritPt | — | 20.9% |
| LMArena Hard Prompts | — | 1465 |
| Epoch Capabilities Index | 156.77 | — |
Math Not comparable
GPT-5.3 Codex: —, Step 5 Preview: 46.1 (#72)
| Benchmark | GPT-5.3 Codex | Step 5 Preview |
|---|---|---|
| ProofBench | — | 42% |
| LMArena Math | — | 1470 |
Knowledge Not comparable
GPT-5.3 Codex: —, Step 5 Preview: 41.2 (#112)
| Benchmark | GPT-5.3 Codex | Step 5 Preview |
|---|---|---|
| LMArena Expert | — | 1470 |
Multimodal Not comparable
GPT-5.3 Codex: —, Step 5 Preview: 41.0 (#41)
| Benchmark | GPT-5.3 Codex | Step 5 Preview |
|---|---|---|
| LMArena Vision | — | 1267 |
Multilingual Not comparable
GPT-5.3 Codex: —, Step 5 Preview: 53.8 (#54)
| Benchmark | GPT-5.3 Codex | Step 5 Preview |
|---|---|---|
| LMArena Non-English | — | 1432 |
| LMArena Chinese | — | 1519 |
| LMArena Russian | — | 1436 |
| LMArena Spanish | — | 1439 |
Instruction Following Not comparable
GPT-5.3 Codex: —, Step 5 Preview: 76.0 (#50)
| Benchmark | GPT-5.3 Codex | Step 5 Preview |
|---|---|---|
| LMArena Instruction Following | — | 1444 |
Long Context Not comparable
GPT-5.3 Codex: —, Step 5 Preview: 44.5 (#64)
| Benchmark | GPT-5.3 Codex | Step 5 Preview |
|---|---|---|
| LMArena Longer Query | — | 1455 |
Writing & Preference Not comparable
GPT-5.3 Codex: —, Step 5 Preview: 62.8 (#71)
| Benchmark | GPT-5.3 Codex | Step 5 Preview |
|---|---|---|
| LMArena Text | — | 1442 |
| LMArena Creative Writing | — | 1410 |
| LMArena Multi-Turn | — | 1447 |
Frequently asked questions
Is GPT-5.3 Codex better than Step 5 Preview?
Step 5 Preview is the stronger model overall, scoring 47.9 to 45.8 on the Noometry Index.
Which is cheaper, GPT-5.3 Codex or Step 5 Preview?
Step 5 Preview is cheaper. It lists at $1 per million input tokens and $2.70 per million output tokens; GPT-5.3 Codex lists at $1.75 and $14.
Is GPT-5.3 Codex or Step 5 Preview better for coding?
Step 5 Preview scores higher on coding benchmarks: 51.8 versus 48.6 in the Noometry coding category.
Which has the bigger context window?
Step 5 Preview does, with 1.02M tokens against 400K.
How many benchmarks do GPT-5.3 Codex and Step 5 Preview share?
1 benchmark has published results for both models. GPT-5.3 Codex has 8 scored results on Noometry and Step 5 Preview has 18.