Model comparison
GPT-5-Codex vs Step 1o Turbo 202506
Step 1o Turbo 202506 is the stronger model overall, scoring 39.7 to 37.9 on the Noometry Index.
Last verified . 0 shared benchmarks.
StepFunSummary
- The widest gap is in reasoning, where GPT-5-Codex leads 30.9 to 26.8.
Side by side
| GPT-5-Codex | Step 1o Turbo 202506 | |
|---|---|---|
| Provider | OpenAI | StepFun |
| Noometry Index | 37.9 | 39.7 |
| Released | 2025-09-15 | — |
| Weights | Proprietary | Proprietary |
| Context window | 400K | — |
| Max output | 128K | — |
| Input $ / M tokens | $1.25 | — |
| Output $ / M tokens | $10 | — |
| Results tracked | 3 | 14 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5-Codex leads
GPT-5-Codex: 42.4 (#103), Step 1o Turbo 202506: 39.2 (#160)
| Benchmark | GPT-5-Codex | Step 1o Turbo 202506 |
|---|---|---|
| WeirdML | 54.5% | — |
| LMArena Coding | — | 1339 |
Agentic & Tool Use Not comparable
GPT-5-Codex: 31.0 (#72), Step 1o Turbo 202506: —
| Benchmark | GPT-5-Codex | Step 1o Turbo 202506 |
|---|---|---|
| Terminal-Bench | 44.3% | — |
Reasoning GPT-5-Codex leads
GPT-5-Codex: 30.9 (#83), Step 1o Turbo 202506: 26.8 (#129)
| Benchmark | GPT-5-Codex | Step 1o Turbo 202506 |
|---|---|---|
| Kagi LLM Benchmark | 70.3% | — |
| LMArena Hard Prompts | — | 1335 |
Math Not comparable
GPT-5-Codex: —, Step 1o Turbo 202506: 36.6 (#164)
| Benchmark | GPT-5-Codex | Step 1o Turbo 202506 |
|---|---|---|
| LMArena Math | — | 1318 |
Knowledge Not comparable
GPT-5-Codex: —, Step 1o Turbo 202506: 36.1 (#176)
| Benchmark | GPT-5-Codex | Step 1o Turbo 202506 |
|---|---|---|
| LMArena Expert | — | 1308 |
Multimodal Not comparable
GPT-5-Codex: —, Step 1o Turbo 202506: 36.1 (#80)
| Benchmark | GPT-5-Codex | Step 1o Turbo 202506 |
|---|---|---|
| LMArena Vision | — | 1186 |
Multilingual Not comparable
GPT-5-Codex: —, Step 1o Turbo 202506: 45.3 (#173)
| Benchmark | GPT-5-Codex | Step 1o Turbo 202506 |
|---|---|---|
| LMArena Non-English | — | 1313 |
| LMArena Chinese | — | 1380 |
| LMArena German | — | 1314 |
| LMArena Russian | — | 1327 |
Instruction Following Not comparable
GPT-5-Codex: —, Step 1o Turbo 202506: 69.2 (#176)
| Benchmark | GPT-5-Codex | Step 1o Turbo 202506 |
|---|---|---|
| LMArena Instruction Following | — | 1310 |
Long Context Not comparable
GPT-5-Codex: —, Step 1o Turbo 202506: 41.0 (#146)
| Benchmark | GPT-5-Codex | Step 1o Turbo 202506 |
|---|---|---|
| LMArena Longer Query | — | 1348 |
Writing & Preference Not comparable
GPT-5-Codex: —, Step 1o Turbo 202506: 53.1 (#160)
| Benchmark | GPT-5-Codex | Step 1o Turbo 202506 |
|---|---|---|
| LMArena Text | — | 1336 |
| LMArena Creative Writing | — | 1307 |
| LMArena Multi-Turn | — | 1340 |
Frequently asked questions
Is GPT-5-Codex better than Step 1o Turbo 202506?
Step 1o Turbo 202506 is the stronger model overall, scoring 39.7 to 37.9 on the Noometry Index.
Is GPT-5-Codex or Step 1o Turbo 202506 better for coding?
GPT-5-Codex scores higher on coding benchmarks: 42.4 versus 39.2 in the Noometry coding category.
How many benchmarks do GPT-5-Codex and Step 1o Turbo 202506 share?
0 benchmarks have published results for both models. GPT-5-Codex has 3 scored results on Noometry and Step 1o Turbo 202506 has 14.