Noometry

Model comparison

GPT-4 vs Step 3.5 Flash

Step 3.5 Flash is the stronger model overall, scoring 42.3 to 29.1 on the Noometry Index.

Last verified . 17 shared benchmarks.

GPT-4 OpenAI

29.1

Rank #316 Confirmed

Step 3.5 Flash StepFun

42.3

Rank #116 Confirmed

Summary

  • They share 17 benchmarks with published results for both. GPT-4 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 10.8.
  • Step 3.5 Flash is cheaper at $0.10 / $0.30 per million input/output tokens, against $30 / $60 for GPT-4.
  • Step 3.5 Flash accepts more context: 256K tokens versus 8K.
  • Step 3.5 Flash has downloadable open weights; the other is API-only.

Side by side

GPT-4Step 3.5 Flash
ProviderOpenAIStepFun
Noometry Index29.142.3
Released2023-03-142026-01-29
WeightsProprietaryOpen
Context window8K256K
Max output8K256K
Input $ / M tokens$30$0.10
Output $ / M tokens$60$0.30
Results tracked3819

Sponsored placements are available on pages like this one. Advertise on Noometry

Category by category

Coding Step 3.5 Flash leads

GPT-4: 31.6 (#283), Step 3.5 Flash: 42.4 (#105)

BenchmarkGPT-4Step 3.5 Flash
LMArena Coding12541436
WeirdML12.4%—
BigCodeBench Instruct46%—
BigCodeBench Complete57.2%—
HumanEval+79.3%—

Agentic & Tool Use Not comparable

GPT-4: —, Step 3.5 Flash: —

BenchmarkGPT-4Step 3.5 Flash
METR Time Horizons36.1%—

Reasoning Step 3.5 Flash leads

GPT-4: 17.8 (#289), Step 3.5 Flash: 22.2 (#202)

BenchmarkGPT-4Step 3.5 Flash
LMArena Hard Prompts12411411
NYT Connections (extended)—28.4%
Chess Puzzles4%—
Mystery Game Puzzles12%—
DTBench62.7%—
LMCA17.1%—
BIG-Bench Hard75.1%—
Epoch Capabilities Index125.89—
ForecastBench57.8—
HellaSwag95.3%—
WinoGrande87.5%—

Math Step 3.5 Flash leads

GPT-4: 10.8 (#309), Step 3.5 Flash: 42.6 (#84)

BenchmarkGPT-4Step 3.5 Flash
LMArena Math12691408
MathArena Final-Answer Competitions—66.8%
OTIS Mock AIME 2024-20251.1%—
MATH Level 523%—
GSM8K92%—

Knowledge Step 3.5 Flash leads

GPT-4: 18.4 (#282), Step 3.5 Flash: 39.6 (#132)

BenchmarkGPT-4Step 3.5 Flash
LMArena Expert12111421
GPQA Diamond35.7%—
MMLU86.4%—
TriviaQA84.8%—

Multilingual Step 3.5 Flash leads

GPT-4: 40.6 (#215), Step 3.5 Flash: 50.5 (#119)

BenchmarkGPT-4Step 3.5 Flash
LMArena Non-English12461385
LMArena Chinese12421447
LMArena French12831421
LMArena German12511405
LMArena Japanese12091354
LMArena Korean11841352
LMArena Russian12511385
LMArena Spanish12611419

Instruction Following Step 3.5 Flash leads

GPT-4: 65.3 (#222), Step 3.5 Flash: 73.1 (#124)

BenchmarkGPT-4Step 3.5 Flash
LMArena Instruction Following12411385

Long Context Step 3.5 Flash leads

GPT-4: 37.7 (#212), Step 3.5 Flash: 42.8 (#117)

BenchmarkGPT-4Step 3.5 Flash
LMArena Longer Query12441402

Writing & Preference Step 3.5 Flash leads

GPT-4: 34.9 (#268), Step 3.5 Flash: 58.8 (#113)

BenchmarkGPT-4Step 3.5 Flash
LMArena Text12631403
LMArena Creative Writing12441357
LMArena Multi-Turn12571405
EQ-Bench Creative Writing752—

Frequently asked questions

Is GPT-4 better than Step 3.5 Flash?

Step 3.5 Flash is the stronger model overall, scoring 42.3 to 29.1 on the Noometry Index.

Which is cheaper, GPT-4 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-4 lists at $30 and $60.

Is GPT-4 or Step 3.5 Flash better for coding?

Step 3.5 Flash scores higher on coding benchmarks: 42.4 versus 31.6 in the Noometry coding category.

Which has the bigger context window?

Step 3.5 Flash does, with 256K tokens against 8K.

How many benchmarks do GPT-4 and Step 3.5 Flash share?

17 benchmarks have published results for both models. GPT-4 has 38 scored results on Noometry and Step 3.5 Flash has 19.

Related comparisons

Go deeper