Model comparison
gpt-oss-120b vs Step 3.5 Flash
Step 3.5 Flash is the stronger model overall, scoring 42.3 to 36.3 on the Noometry Index. gpt-oss-120b costs 2.1× less per token, which makes it the better buy when Step 3.5 Flash's lead doesn't matter for your workload.
Last verified . 17 shared benchmarks.
StepFunSummary
- They share 17 benchmarks with published results for both. gpt-oss-120b scores higher in 2 categories and Step 3.5 Flash in 6 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Step 3.5 Flash leads 58.8 to 46.5.
- gpt-oss-120b is cheaper at $0.037 / $0.17 per million input/output tokens, against $0.10 / $0.30 for Step 3.5 Flash.
- Step 3.5 Flash accepts more context: 256K tokens versus 131K.
Side by side
| gpt-oss-120b | Step 3.5 Flash | |
|---|---|---|
| Provider | OpenAI | StepFun |
| Noometry Index | 36.3 | 42.3 |
| Released | 2025-08-05 | 2026-01-29 |
| Weights | Open | Open |
| Context window | 131K | 256K |
| Max output | 41K | 256K |
| Input $ / M tokens | $0.037 | $0.10 |
| Output $ / M tokens | $0.17 | $0.30 |
| Results tracked | 48 | 19 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Step 3.5 Flash leads
gpt-oss-120b: 33.5 (#256), Step 3.5 Flash: 42.4 (#105)
| Benchmark | gpt-oss-120b | Step 3.5 Flash |
|---|---|---|
| LMArena Coding | 1380 | 1436 |
| SWE-bench Verified (bash only) | 26% | — |
| Aider Polyglot | 41.8% | — |
| SciCode | 36% | — |
| WeirdML | 48.2% | — |
| ALE-Bench | 575.62 | — |
| AlgoTune | 1.41 | — |
Agentic & Tool Use Not comparable
gpt-oss-120b: 12.2 (#153), Step 3.5 Flash: —
| Benchmark | gpt-oss-120b | Step 3.5 Flash |
|---|---|---|
| Terminal-Bench | 18.7% | — |
| APEX-Agents | 4.4% | — |
| METR Time Horizons | 56.6% | — |
| Vending-Bench 2 | -21.53 | — |
Reasoning Step 3.5 Flash leads
gpt-oss-120b: 20.0 (#245), Step 3.5 Flash: 22.2 (#202)
| Benchmark | gpt-oss-120b | Step 3.5 Flash |
|---|---|---|
| LMArena Hard Prompts | 1364 | 1411 |
| SimpleBench | 22.1% | — |
| Kagi LLM Benchmark | 58.6% | — |
| NYT Connections (extended) | — | 28.4% |
| CritPt | 1.1% | — |
| Chess Puzzles | 20% | — |
| Mystery Game Puzzles | 2% | — |
| DTBench | 76.3% | — |
| LMCA | 22.1% | — |
| Surface Evolver Bench | 25% | — |
| Epoch Capabilities Index | 139.93 | — |
Math gpt-oss-120b leads
gpt-oss-120b: 52.5 (#50), Step 3.5 Flash: 42.6 (#84)
| Benchmark | gpt-oss-120b | Step 3.5 Flash |
|---|---|---|
| LMArena Math | 1389 | 1408 |
| MathArena Final-Answer Competitions | — | 66.8% |
| OTIS Mock AIME 2024-2025 | 88.9% | — |
| Omni-MATH | 68.8% | — |
Knowledge gpt-oss-120b leads
gpt-oss-120b: 42.4 (#96), Step 3.5 Flash: 39.6 (#132)
| Benchmark | gpt-oss-120b | Step 3.5 Flash |
|---|---|---|
| LMArena Expert | 1356 | 1421 |
| GPQA Diamond | 75.8% | — |
| MMLU-Pro | 79.5% | — |
| Confabulations | 15.7% | — |
| Vectara Hallucination Rate | 14.2% | — |
| GPQA (HELM) | 68.4% | — |
Multilingual Step 3.5 Flash leads
gpt-oss-120b: 48.0 (#147), Step 3.5 Flash: 50.5 (#119)
| Benchmark | gpt-oss-120b | Step 3.5 Flash |
|---|---|---|
| LMArena Non-English | 1351 | 1385 |
| LMArena Chinese | 1385 | 1447 |
| LMArena French | 1369 | 1421 |
| LMArena German | 1353 | 1405 |
| LMArena Japanese | 1331 | 1354 |
| LMArena Korean | 1282 | 1352 |
| LMArena Russian | 1343 | 1385 |
| LMArena Spanish | 1389 | 1419 |
Instruction Following Step 3.5 Flash leads
gpt-oss-120b: 69.3 (#173), Step 3.5 Flash: 73.1 (#124)
| Benchmark | gpt-oss-120b | Step 3.5 Flash |
|---|---|---|
| LMArena Instruction Following | 1318 | 1385 |
| IFEval | 83.6% | — |
Long Context Step 3.5 Flash leads
gpt-oss-120b: 31.4 (#278), Step 3.5 Flash: 42.8 (#117)
| Benchmark | gpt-oss-120b | Step 3.5 Flash |
|---|---|---|
| LMArena Longer Query | 1319 | 1402 |
| Fiction.LiveBench | 44.4% | — |
Writing & Preference Step 3.5 Flash leads
gpt-oss-120b: 46.5 (#217), Step 3.5 Flash: 58.8 (#113)
| Benchmark | gpt-oss-120b | Step 3.5 Flash |
|---|---|---|
| LMArena Text | 1365 | 1403 |
| LMArena Creative Writing | 1275 | 1357 |
| LMArena Multi-Turn | 1340 | 1405 |
| Short-Story Creative Writing | 77.1% | — |
| EQ-Bench Creative Writing | 961 | — |
| WildBench | 84.5% | — |
Frequently asked questions
Is gpt-oss-120b better than Step 3.5 Flash?
Step 3.5 Flash is the stronger model overall, scoring 42.3 to 36.3 on the Noometry Index. gpt-oss-120b costs 2.1× less per token, which makes it the better buy when Step 3.5 Flash's lead doesn't matter for your workload.
Which is cheaper, gpt-oss-120b or Step 3.5 Flash?
gpt-oss-120b is cheaper. It lists at $0.037 per million input tokens and $0.17 per million output tokens; Step 3.5 Flash lists at $0.10 and $0.30.
Is gpt-oss-120b or Step 3.5 Flash better for coding?
Step 3.5 Flash scores higher on coding benchmarks: 42.4 versus 33.5 in the Noometry coding category.
Which has the bigger context window?
Step 3.5 Flash does, with 256K tokens against 131K.
How many benchmarks do gpt-oss-120b and Step 3.5 Flash share?
17 benchmarks have published results for both models. gpt-oss-120b has 48 scored results on Noometry and Step 3.5 Flash has 19.