Model comparison
GPT-3.5-turbo vs Step 3.5 Flash
Step 3.5 Flash is the stronger model overall, scoring 42.3 to 23.2 on the Noometry Index.
Last verified . 17 shared benchmarks.
StepFunSummary
- They share 17 benchmarks with published results for both. GPT-3.5-turbo scores higher in 0 categories and Step 3.5 Flash in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Step 3.5 Flash leads 42.6 to 6.3.
- Step 3.5 Flash is cheaper at $0.10 / $0.30 per million input/output tokens, against $0.50 / $1.50 for GPT-3.5-turbo.
- Step 3.5 Flash accepts more context: 256K tokens versus 16K.
- Step 3.5 Flash has downloadable open weights; the other is API-only.
Side by side
| GPT-3.5-turbo | Step 3.5 Flash | |
|---|---|---|
| Provider | OpenAI | StepFun |
| Noometry Index | 23.2 | 42.3 |
| Released | 2023-03-01 | 2026-01-29 |
| Weights | Proprietary | Open |
| Context window | 16K | 256K |
| Max output | 4K | 256K |
| Input $ / M tokens | $0.50 | $0.10 |
| Output $ / M tokens | $1.50 | $0.30 |
| Results tracked | 44 | 19 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Step 3.5 Flash leads
GPT-3.5-turbo: 23.9 (#331), Step 3.5 Flash: 42.4 (#105)
| Benchmark | GPT-3.5-turbo | Step 3.5 Flash |
|---|---|---|
| LMArena Coding | 1136 | 1436 |
| WeirdML | 3.5% | — |
| BigCodeBench Instruct | 39.1% | — |
| BigCodeBench Complete | 50.6% | — |
| HumanEval+ | 70.7% | — |
| MBPP+ | 69.7% | — |
Agentic & Tool Use Not comparable
GPT-3.5-turbo: —, Step 3.5 Flash: —
| Benchmark | GPT-3.5-turbo | Step 3.5 Flash |
|---|---|---|
| METR Time Horizons | 21.5% | — |
Reasoning Step 3.5 Flash leads
GPT-3.5-turbo: 13.8 (#332), Step 3.5 Flash: 22.2 (#202)
| Benchmark | GPT-3.5-turbo | Step 3.5 Flash |
|---|---|---|
| LMArena Hard Prompts | 1108 | 1411 |
| NYT Connections (extended) | — | 28.4% |
| Chess Puzzles | 0% | — |
| Mystery Game Puzzles | 3% | — |
| DTBench | 48.5% | — |
| LMCA | 9.7% | — |
| Adversarial NLI | 58.1% | — |
| BIG-Bench Hard | 61.6% | — |
| CommonsenseQA 2.0 | 57% | — |
| Epoch Capabilities Index | 118.55 | — |
| ForecastBench | 50.4 | — |
| WinoGrande | 81.6% | — |
Math Step 3.5 Flash leads
GPT-3.5-turbo: 6.3 (#327), Step 3.5 Flash: 42.6 (#84)
| Benchmark | GPT-3.5-turbo | Step 3.5 Flash |
|---|---|---|
| LMArena Math | 1142 | 1408 |
| FrontierMath (Tiers 1-3) | 0% | — |
| MathArena Final-Answer Competitions | — | 66.8% |
| OTIS Mock AIME 2024-2025 | 2.2% | — |
| MATH Level 5 | 15.9% | — |
| GSM8K | 57.8% | — |
Knowledge Step 3.5 Flash leads
GPT-3.5-turbo: 10.0 (#303), Step 3.5 Flash: 39.6 (#132)
| Benchmark | GPT-3.5-turbo | Step 3.5 Flash |
|---|---|---|
| LMArena Expert | 1070 | 1421 |
| GPQA Diamond | 28% | — |
| ARC (AI2) Challenge | 87.4% | — |
| BoolQ | 87% | — |
| MMLU | 71.4% | — |
| OpenBookQA | 86% | — |
| TriviaQA | 85.8% | — |
Multilingual Step 3.5 Flash leads
GPT-3.5-turbo: 31.5 (#258), Step 3.5 Flash: 50.5 (#119)
| Benchmark | GPT-3.5-turbo | Step 3.5 Flash |
|---|---|---|
| LMArena Non-English | 1108 | 1385 |
| LMArena Chinese | 1075 | 1447 |
| LMArena French | 1118 | 1421 |
| LMArena German | 1090 | 1405 |
| LMArena Japanese | 1043 | 1354 |
| LMArena Korean | 1019 | 1352 |
| LMArena Russian | 1123 | 1385 |
| LMArena Spanish | 1121 | 1419 |
Instruction Following Step 3.5 Flash leads
GPT-3.5-turbo: 57.9 (#262), Step 3.5 Flash: 73.1 (#124)
| Benchmark | GPT-3.5-turbo | Step 3.5 Flash |
|---|---|---|
| LMArena Instruction Following | 1119 | 1385 |
Long Context Step 3.5 Flash leads
GPT-3.5-turbo: 34.0 (#254), Step 3.5 Flash: 42.8 (#117)
| Benchmark | GPT-3.5-turbo | Step 3.5 Flash |
|---|---|---|
| LMArena Longer Query | 1121 | 1402 |
Writing & Preference Step 3.5 Flash leads
GPT-3.5-turbo: 25.3 (#305), Step 3.5 Flash: 58.8 (#113)
| Benchmark | GPT-3.5-turbo | Step 3.5 Flash |
|---|---|---|
| LMArena Text | 1125 | 1403 |
| LMArena Creative Writing | 1092 | 1357 |
| LMArena Multi-Turn | 1117 | 1405 |
| EQ-Bench Creative Writing | 451 | — |
Frequently asked questions
Is GPT-3.5-turbo better than Step 3.5 Flash?
Step 3.5 Flash is the stronger model overall, scoring 42.3 to 23.2 on the Noometry Index.
Which is cheaper, GPT-3.5-turbo or Step 3.5 Flash?
Step 3.5 Flash is cheaper. It lists at $0.10 per million input tokens and $0.30 per million output tokens; GPT-3.5-turbo lists at $0.50 and $1.50.
Is GPT-3.5-turbo or Step 3.5 Flash better for coding?
Step 3.5 Flash scores higher on coding benchmarks: 42.4 versus 23.9 in the Noometry coding category.
Which has the bigger context window?
Step 3.5 Flash does, with 256K tokens against 16K.
How many benchmarks do GPT-3.5-turbo and Step 3.5 Flash share?
17 benchmarks have published results for both models. GPT-3.5-turbo has 44 scored results on Noometry and Step 3.5 Flash has 19.