Noometry

Model comparison

Claude 3 Haiku vs gpt-oss-20b

gpt-oss-20b is the stronger model overall, scoring 32.1 to 25.7 on the Noometry Index.

Last verified . 24 shared benchmarks.

Claude 3 Haiku Anthropic

25.7

Rank #333 Confirmed

gpt-oss-20b OpenAI

32.1

Rank #249 Confirmed

Summary

  • They share 24 benchmarks with published results for both. Claude 3 Haiku scores higher in 0 categories and gpt-oss-20b in 8 categories; 7 gaps are clear of the uncertainty.
  • The widest gap is in math, where gpt-oss-20b leads 37.6 to 9.2.
  • The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 1.8% for Claude 3 Haiku and 65.3% for gpt-oss-20b.
  • gpt-oss-20b has downloadable open weights; the other is API-only.

Side by side

Claude 3 Haikugpt-oss-20b
ProviderAnthropicOpenAI
Noometry Index25.732.1
Released2024-03-072025-08-05
WeightsProprietaryOpen
Context window—131K
Max output—16K
Input $ / M tokens—$0.018
Output $ / M tokens—$0.09
Results tracked3734

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

Category by category

Coding gpt-oss-20b leads

Claude 3 Haiku: 26.1 (#320), gpt-oss-20b: 37.1 (#185)

BenchmarkClaude 3 Haikugpt-oss-20b
WeirdML9.8%40.9%
LMArena Coding11991307
SciCode—34.4%
BigCodeBench Instruct39.4%—
BigCodeBench Complete50.1%—
CadEval12%—
ALE-Bench—566.05
HumanEval+68.9%—
MBPP+68.8%—

Agentic & Tool Use Not comparable

Claude 3 Haiku: —, gpt-oss-20b: 9.2 (#146)

BenchmarkClaude 3 Haikugpt-oss-20b
Terminal-Bench—3.4%

Reasoning gpt-oss-20b leads

Claude 3 Haiku: 15.7 (#303), gpt-oss-20b: 18.8 (#251)

BenchmarkClaude 3 Haikugpt-oss-20b
Kagi LLM Benchmark34.2%53.2%
LMArena Hard Prompts11741274
DTBench50.1%68%
LMCA8.8%14.5%
Epoch Capabilities Index118.3137.81
CritPt—1.4%
Chess Puzzles—4%
ForecastBench53.2—
WinoGrande74.2%—

Math gpt-oss-20b leads

Claude 3 Haiku: 9.2 (#313), gpt-oss-20b: 37.6 (#93)

BenchmarkClaude 3 Haikugpt-oss-20b
OTIS Mock AIME 2024-20251.8%65.3%
LMArena Math11881317
Omni-MATH—56.5%
MATH Level 514.9%—

Knowledge gpt-oss-20b leads

Claude 3 Haiku: 17.3 (#280), gpt-oss-20b: 34.6 (#186)

BenchmarkClaude 3 Haikugpt-oss-20b
GPQA Diamond36.3%60.8%
LMArena Expert11481258
MMLU-Pro—74%
Confabulations34.2%—
GPQA (HELM)—59.4%
MMLU73.8%—

Multimodal Not comparable

Claude 3 Haiku: 23.1 (#123), gpt-oss-20b: —

BenchmarkClaude 3 Haikugpt-oss-20b
LMArena Vision950—
ScienceQA72%—

Multilingual gpt-oss-20b leads

Claude 3 Haiku: 36.1 (#238), gpt-oss-20b: 42.4 (#192)

BenchmarkClaude 3 Haikugpt-oss-20b
LMArena Non-English11781269
LMArena Chinese11551311
LMArena German11741249
LMArena Japanese11021245
LMArena Korean11081239
LMArena Russian12041279
LMArena Spanish11651267
LMArena French1194—

Instruction Following Too close to call

Claude 3 Haiku: 61.5 (#242), gpt-oss-20b: 62.0 (#235)

BenchmarkClaude 3 Haikugpt-oss-20b
LMArena Instruction Following11741236
IFEval—73.2%

Long Context gpt-oss-20b leads

Claude 3 Haiku: 36.1 (#232), gpt-oss-20b: 37.9 (#203)

BenchmarkClaude 3 Haikugpt-oss-20b
LMArena Longer Query11911251

Writing & Preference gpt-oss-20b leads

Claude 3 Haiku: 29.8 (#286), gpt-oss-20b: 35.8 (#260)

BenchmarkClaude 3 Haikugpt-oss-20b
LMArena Text11951288
LMArena Creative Writing11571201
EQ-Bench Creative Writing717666
LMArena Multi-Turn11901268
WildBench—73.7%

Frequently asked questions

Is Claude 3 Haiku better than gpt-oss-20b?

gpt-oss-20b is the stronger model overall, scoring 32.1 to 25.7 on the Noometry Index.

Is Claude 3 Haiku or gpt-oss-20b better for coding?

gpt-oss-20b scores higher on coding benchmarks: 37.1 versus 26.1 in the Noometry coding category.

How many benchmarks do Claude 3 Haiku and gpt-oss-20b share?

24 benchmarks have published results for both models. Claude 3 Haiku has 37 scored results on Noometry and gpt-oss-20b has 34.

Related comparisons

Go deeper