Noometry

Model comparison

Claude 3 Opus vs Olmo 3.1 32b Instruct

Olmo 3.1 32b Instruct is the stronger model overall, scoring 38.8 to 29.3 on the Noometry Index.

Last verified . 16 shared benchmarks.

Claude 3 Opus Anthropic

29.3

Rank #300 Confirmed

Olmo 3.1 32b Instruct Allen Institute for AI (Ai2)

38.8

Rank #168 Confirmed

Summary

  • They share 16 benchmarks with published results for both. Claude 3 Opus scores higher in 0 categories and Olmo 3.1 32b Instruct in 8 categories; 8 gaps are clear of the uncertainty.
  • The widest gap is in math, where Olmo 3.1 32b Instruct leads 33.6 to 13.9.
  • Olmo 3.1 32b Instruct has downloadable open weights; the other is API-only.

Side by side

Claude 3 OpusOlmo 3.1 32b Instruct
ProviderAnthropicAllen Institute for AI (Ai2)
Noometry Index29.338.8
Released2024-02-29—
WeightsProprietaryOpen
Context window——
Max output——
Input $ / M tokens——
Output $ / M tokens——
Results tracked4616

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

Category by category

Coding Olmo 3.1 32b Instruct leads

Claude 3 Opus: 32.4 (#261), Olmo 3.1 32b Instruct: 38.9 (#154)

BenchmarkClaude 3 OpusOlmo 3.1 32b Instruct
LMArena Coding12641346
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: 24.3 (#110), Olmo 3.1 32b Instruct: —

BenchmarkClaude 3 OpusOlmo 3.1 32b Instruct
Cybench10%—
METR Time Horizons29.5%—

Reasoning Olmo 3.1 32b Instruct leads

Claude 3 Opus: 14.2 (#318), Olmo 3.1 32b Instruct: 25.4 (#131)

BenchmarkClaude 3 OpusOlmo 3.1 32b Instruct
LMArena Hard Prompts12461322
SimpleBench23.5%—
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 Olmo 3.1 32b Instruct leads

Claude 3 Opus: 13.9 (#293), Olmo 3.1 32b Instruct: 33.6 (#169)

BenchmarkClaude 3 OpusOlmo 3.1 32b Instruct
LMArena Math12731305
OTIS Mock AIME 2024-20254.7%—
LiveBench Math43.6%—
MATH Level 537.5%—

Knowledge Olmo 3.1 32b Instruct leads

Claude 3 Opus: 24.5 (#262), Olmo 3.1 32b Instruct: 35.4 (#175)

BenchmarkClaude 3 OpusOlmo 3.1 32b Instruct
LMArena Expert12231307
GPQA Diamond47.2%—
SimpleQA Verified12.6%—
Confabulations22.7%—
MMLU84.6%—

Multimodal Not comparable

Claude 3 Opus: 26.6 (#112), Olmo 3.1 32b Instruct: —

BenchmarkClaude 3 OpusOlmo 3.1 32b Instruct
LMArena Vision1023—

Multilingual Olmo 3.1 32b Instruct leads

Claude 3 Opus: 41.6 (#202), Olmo 3.1 32b Instruct: 42.8 (#185)

BenchmarkClaude 3 OpusOlmo 3.1 32b Instruct
LMArena Non-English12581275
LMArena Chinese12481304
LMArena French12741328
LMArena German12581283
LMArena Korean11871206
LMArena Russian12801268
LMArena Spanish12461337
LMArena Japanese1204—

Instruction Following Olmo 3.1 32b Instruct leads

Claude 3 Opus: 64.3 (#223), Olmo 3.1 32b Instruct: 68.8 (#182)

BenchmarkClaude 3 OpusOlmo 3.1 32b Instruct
LMArena Instruction Following12481298
LiveBench Instruction Following63.9%—

Long Context Olmo 3.1 32b Instruct leads

Claude 3 Opus: 38.2 (#197), Olmo 3.1 32b Instruct: 39.9 (#161)

BenchmarkClaude 3 OpusOlmo 3.1 32b Instruct
LMArena Longer Query12591312

Writing & Preference Olmo 3.1 32b Instruct leads

Claude 3 Opus: 47.4 (#208), Olmo 3.1 32b Instruct: 50.4 (#179)

BenchmarkClaude 3 OpusOlmo 3.1 32b Instruct
LMArena Text12621311
LMArena Creative Writing12351264
LMArena Multi-Turn12751309
LiveBench Language50.4%—

Frequently asked questions

Is Claude 3 Opus better than Olmo 3.1 32b Instruct?

Olmo 3.1 32b Instruct is the stronger model overall, scoring 38.8 to 29.3 on the Noometry Index.

Is Claude 3 Opus or Olmo 3.1 32b Instruct better for coding?

Olmo 3.1 32b Instruct scores higher on coding benchmarks: 38.9 versus 32.4 in the Noometry coding category.

How many benchmarks do Claude 3 Opus and Olmo 3.1 32b Instruct share?

16 benchmarks have published results for both models. Claude 3 Opus has 46 scored results on Noometry and Olmo 3.1 32b Instruct has 16.

Related comparisons

Go deeper