Model comparison

Claude 3 Opus vs Step 3

Step 3 is the stronger model overall, scoring 40.5 to 29.5 on the Noometry Index.

Last verified . 16 shared benchmarks.

Claude 3 Opus Anthropic

29.5

Rank #310 Confirmed

Step 3 StepFun

40.5

Rank #149 Confirmed

Summary

  • They share 16 benchmarks with published results for both. Claude 3 Opus scores higher in 0 categories and Step 3 in 9 categories; 9 gaps are clear of the uncertainty.
  • The widest gap is in math, where Step 3 leads 37.6 to 14.8.
  • Step 3 has downloadable open weights; the other is API-only.

Side by side

Claude 3 Opus and Step 3 specifications
Claude 3 OpusStep 3
ProviderAnthropicStepFun
Noometry Index29.540.5
Released2024-02-29—
WeightsProprietaryOpen
Context window——
Max output——
Input $ / M tokens——
Output $ / M tokens——
Results tracked4617

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

Category by category

Coding Step 3 leads

Claude 3 Opus: 32.9 (#267), Step 3: 40.1 (#147)

Coding benchmarks
BenchmarkClaude 3 OpusStep 3
LMArena Coding12641367
WeirdML19.2%—
BigCodeBench Instruct45.5%—
LiveBench Coding38.6%—
BigCodeBench Complete57.4%—
HumanEval+77.4%—
MBPP+73.3%—

Agentic & Tool Use Not comparable

Claude 3 Opus: 24.6 (#116), Step 3: —

Agentic & Tool Use benchmarks
BenchmarkClaude 3 OpusStep 3
Cybench10%—
METR Time Horizons29.5%—

Reasoning Step 3 leads

Claude 3 Opus: 14.6 (#324), Step 3: 28.4 (#105)

Reasoning benchmarks
BenchmarkClaude 3 OpusStep 3
LMArena Hard Prompts12451355
SimpleBench23.5%—
Kagi LLM Benchmark—62.3%
Chess Puzzles5%—
EnigmaEval0.8%—
LiveBench Reasoning40.6%—
DTBench61.6%—
LiveBench Data Analysis57.9%—
LMCA17%—
Epoch Capabilities Index126.91—
ForecastBench58.4—
LiveBench49.2%—
WinoGrande88.5%—

Math Step 3 leads

Claude 3 Opus: 14.8 (#299), Step 3: 37.6 (#148)

Math benchmarks
BenchmarkClaude 3 OpusStep 3
LMArena Math12731366
OTIS Mock AIME 2024-20254.7%—
LiveBench Math43.6%—
MATH Level 537.5%—

Knowledge Step 3 leads

Claude 3 Opus: 24.5 (#267), Step 3: 36.8 (#164)

Knowledge benchmarks
BenchmarkClaude 3 OpusStep 3
LMArena Expert12231333
GPQA Diamond47.2%—
SimpleQA Verified12.6%—
Confabulations22.7%—
MMLU84.6%—

Multimodal Step 3 leads

Claude 3 Opus: 27.1 (#116), Step 3: 35.5 (#86)

Multimodal benchmarks
BenchmarkClaude 3 OpusStep 3
LMArena Vision10231177

Multilingual Step 3 leads

Claude 3 Opus: 41.4 (#207), Step 3: 46.3 (#159)

Multilingual benchmarks
BenchmarkClaude 3 OpusStep 3
LMArena Non-English12581327
LMArena Chinese12481397
LMArena German12581371
LMArena Korean11871269
LMArena Russian12801331
LMArena Spanish12461371
LMArena French1275—
LMArena Japanese1204—

Instruction Following Step 3 leads

Claude 3 Opus: 64.1 (#228), Step 3: 70.4 (#164)

Instruction Following benchmarks
BenchmarkClaude 3 OpusStep 3
LMArena Instruction Following12481332
LiveBench Instruction Following63.9%—

Long Context Step 3 leads

Claude 3 Opus: 38.2 (#202), Step 3: 40.3 (#157)

Long Context benchmarks
BenchmarkClaude 3 OpusStep 3
LMArena Longer Query12591326

Writing & Preference Step 3 leads

Claude 3 Opus: 47.2 (#213), Step 3: 54.3 (#151)

Writing & Preference benchmarks
BenchmarkClaude 3 OpusStep 3
LMArena Text12621350
LMArena Creative Writing12351321
LMArena Multi-Turn12751341
LiveBench Language50.4%—

Frequently asked questions

Is Claude 3 Opus better than Step 3?

Step 3 is the stronger model overall, scoring 40.5 to 29.5 on the Noometry Index.

Is Claude 3 Opus or Step 3 better for coding?

Step 3 scores higher on coding benchmarks: 40.1 versus 32.9 in the Noometry coding category.

How many benchmarks do Claude 3 Opus and Step 3 share?

16 benchmarks have published results for both models. Claude 3 Opus has 46 scored results on Noometry and Step 3 has 17.

Related comparisons

Go deeper