Noometry

Model comparison

gpt-oss-20b vs Qwen2.5 7B Instruct

gpt-oss-20b is the stronger model overall, scoring 32.5 to 29.0 on the Noometry Index.

Last verified . 11 shared benchmarks.

gpt-oss-20b OpenAI

32.5

Rank #255 Confirmed

Qwen2.5 7B Instruct Alibaba (Qwen)

29.0

Rank #320 Confirmed

Summary

  • They share 11 benchmarks with published results for both. gpt-oss-20b scores higher in 4 categories and Qwen2.5 7B Instruct in 3 categories; 7 gaps are clear of the uncertainty.
  • The widest gap is in math, where gpt-oss-20b leads 39.4 to 12.6.
  • The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 65.3% for gpt-oss-20b and 2.5% for Qwen2.5 7B Instruct.
  • gpt-oss-20b is cheaper at $0.018 / $0.09 per million input/output tokens, against $0.17 / $0.70 for Qwen2.5 7B Instruct.

Side by side

gpt-oss-20bQwen2.5 7B Instruct
ProviderOpenAIAlibaba (Qwen)
Noometry Index32.529.0
Released2025-08-052024-09
WeightsOpenOpen
Context window131K131K
Max output16K8K
Input $ / M tokens$0.018$0.17
Output $ / M tokens$0.09$0.70
Results tracked3415

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

Category by category

Coding gpt-oss-20b leads

gpt-oss-20b: 37.6 (#192), Qwen2.5 7B Instruct: 36.5 (#208)

Benchmarkgpt-oss-20bQwen2.5 7B Instruct
SciCode34.4%—
WeirdML40.9%—
BigCodeBench Instruct—37.6%
LMArena Coding1306—
BigCodeBench Complete—46.1%
ALE-Bench566.05—

Agentic & Tool Use Qwen2.5 7B Instruct leads

gpt-oss-20b: 9.3 (#154), Qwen2.5 7B Instruct: 23.8 (#124)

Benchmarkgpt-oss-20bQwen2.5 7B Instruct
Terminal-Bench3.4%—
BALROG—7.8%

Reasoning gpt-oss-20b leads

gpt-oss-20b: 19.3 (#261), Qwen2.5 7B Instruct: 14.8 (#322)

Benchmarkgpt-oss-20bQwen2.5 7B Instruct
Chess Puzzles4%0%
DTBench68%47.7%
LMCA14.5%6.4%
Epoch Capabilities Index137.82118.51
Kagi LLM Benchmark53.2%—
CritPt1.4%—
LMArena Hard Prompts1274—

Math gpt-oss-20b leads

gpt-oss-20b: 39.4 (#103), Qwen2.5 7B Instruct: 12.6 (#306)

Benchmarkgpt-oss-20bQwen2.5 7B Instruct
OTIS Mock AIME 2024-202565.3%2.5%
Omni-MATH56.5%29.4%
LMArena Math1317—

Knowledge gpt-oss-20b leads

gpt-oss-20b: 34.6 (#195), Qwen2.5 7B Instruct: 17.0 (#286)

Benchmarkgpt-oss-20bQwen2.5 7B Instruct
GPQA Diamond60.8%35.5%
MMLU-Pro74%53.9%
GPQA (HELM)59.4%34.1%
LMArena Expert1258—
MMLU—72.9%

Multilingual Not comparable

gpt-oss-20b: 42.2 (#197), Qwen2.5 7B Instruct: —

Benchmarkgpt-oss-20bQwen2.5 7B Instruct
LMArena Non-English1268—
LMArena Chinese1314—
LMArena German1255—
LMArena Japanese1244—
LMArena Korean1236—
LMArena Russian1278—
LMArena Spanish1267—

Instruction Following Qwen2.5 7B Instruct leads

gpt-oss-20b: 61.8 (#240), Qwen2.5 7B Instruct: 63.2 (#231)

Benchmarkgpt-oss-20bQwen2.5 7B Instruct
IFEval73.2%74.1%
LMArena Instruction Following1236—

Long Context Not comparable

gpt-oss-20b: 37.9 (#209), Qwen2.5 7B Instruct: —

Benchmarkgpt-oss-20bQwen2.5 7B Instruct
LMArena Longer Query1250—

Writing & Preference Qwen2.5 7B Instruct leads

gpt-oss-20b: 35.5 (#265), Qwen2.5 7B Instruct: 48.8 (#195)

Benchmarkgpt-oss-20bQwen2.5 7B Instruct
WildBench73.7%73.1%
LMArena Text1287—
LMArena Creative Writing1201—
EQ-Bench Creative Writing666—
LMArena Multi-Turn1268—

Frequently asked questions

Is gpt-oss-20b better than Qwen2.5 7B Instruct?

gpt-oss-20b is the stronger model overall, scoring 32.5 to 29.0 on the Noometry Index.

Which is cheaper, gpt-oss-20b or Qwen2.5 7B Instruct?

gpt-oss-20b is cheaper. It lists at $0.018 per million input tokens and $0.09 per million output tokens; Qwen2.5 7B Instruct lists at $0.17 and $0.70.

Is gpt-oss-20b or Qwen2.5 7B Instruct better for coding?

gpt-oss-20b scores higher on coding benchmarks: 37.6 versus 36.5 in the Noometry coding category.

Which has the bigger context window?

Both accept 131K tokens.

How many benchmarks do gpt-oss-20b and Qwen2.5 7B Instruct share?

11 benchmarks have published results for both models. gpt-oss-20b has 34 scored results on Noometry and Qwen2.5 7B Instruct has 15.

Related comparisons

Go deeper