Model comparison

DeepSeek-R1-Distill-Qwen-32B vs Qwen2-72B

DeepSeek-R1-Distill-Qwen-32B is the stronger model overall, scoring 35.5 to 30.0 on the Noometry Index.

Last verified October 10, 2026. 4 shared benchmarks.

DeepSeek-R1-Distill-Qwen-32B

DeepSeek

35.5

Rank #226 Confirmed

Qwen2-72B

Alibaba (Qwen)

30.0

Rank #300 Confirmed

Summary

They share 4 benchmarks with published results for both. DeepSeek-R1-Distill-Qwen-32B scores higher in 5 categories and Qwen2-72B in 2 categories; 6 gaps are clear of the uncertainty.
The widest gap is in knowledge, where DeepSeek-R1-Distill-Qwen-32B leads 35.7 to 21.2.
The biggest single-benchmark swing is GPQA Diamond: 64.1% for DeepSeek-R1-Distill-Qwen-32B and 40.8% for Qwen2-72B.

Side by side

DeepSeek-R1-Distill-Qwen-32B and Qwen2-72B specifications
	DeepSeek-R1-Distill-Qwen-32B	Qwen2-72B
Provider	DeepSeek	Alibaba (Qwen)
Noometry Index	35.5	30.0
Released	2025-01-20	2024-06-07
Weights	Open	Open
Context window	—	—
Max output	—	—
Input $ / M tokens	—	—
Output $ / M tokens	—	—
Results tracked	14	26

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

Category by category

Coding DeepSeek-R1-Distill-Qwen-32B leads

DeepSeek-R1-Distill-Qwen-32B: 36.1 (#212), Qwen2-72B: 29.1 (#310)

Coding benchmarks
Benchmark	DeepSeek-R1-Distill-Qwen-32B	Qwen2-72B
BigCodeBench Instruct	43.9%	38.5%
BigCodeBench Complete	54.9%	54%
WeirdML	—	11.3%
LiveBench Coding	33.7%	—
LMArena Coding	—	1196

Agentic & Tool Use DeepSeek-R1-Distill-Qwen-32B leads

DeepSeek-R1-Distill-Qwen-32B: 28.1 (#94), Qwen2-72B: 17.0 (#146)

Agentic & Tool Use benchmarks
Benchmark	DeepSeek-R1-Distill-Qwen-32B	Qwen2-72B
TheAgentCompany	—	1.1%
BALROG	19.5%	—
METR Time Horizons	—	29.9%

Reasoning Qwen2-72B leads

DeepSeek-R1-Distill-Qwen-32B: 18.2 (#284), Qwen2-72B: 23.2 (#181)

Reasoning benchmarks
Benchmark	DeepSeek-R1-Distill-Qwen-32B	Qwen2-72B
Epoch Capabilities Index	137.44	125.28
Chess Puzzles	1%	—
LiveBench Reasoning	52.3%	—
LMArena Hard Prompts	—	1191
LiveBench Data Analysis	45.4%	—
LiveBench	45.5%	—

Math DeepSeek-R1-Distill-Qwen-32B leads

DeepSeek-R1-Distill-Qwen-32B: 34.5 (#194), Qwen2-72B: 30.2 (#236)

Math benchmarks
Benchmark	DeepSeek-R1-Distill-Qwen-32B	Qwen2-72B
OTIS Mock AIME 2024-2025	55.6%	—
LiveBench Math	59.4%	—
LMArena Math	—	1235
MATH Level 5	—	39.1%

Knowledge DeepSeek-R1-Distill-Qwen-32B leads

DeepSeek-R1-Distill-Qwen-32B: 35.7 (#182), Qwen2-72B: 21.2 (#275)

Knowledge benchmarks
Benchmark	DeepSeek-R1-Distill-Qwen-32B	Qwen2-72B
GPQA Diamond	64.1%	40.8%
LMArena Expert	—	1171
MMLU	—	82.4%

Multilingual Not comparable

DeepSeek-R1-Distill-Qwen-32B: —, Qwen2-72B: 35.9 (#244)

Multilingual benchmarks
Benchmark	DeepSeek-R1-Distill-Qwen-32B	Qwen2-72B
LMArena Non-English	—	1176
LMArena Chinese	—	1240
LMArena French	—	1170
LMArena German	—	1151
LMArena Japanese	—	1111
LMArena Korean	—	1083
LMArena Russian	—	1169
LMArena Spanish	—	1169

Instruction Following Too close to call

DeepSeek-R1-Distill-Qwen-32B: 61.6 (#243), Qwen2-72B: 61.7 (#241)

Instruction Following benchmarks
Benchmark	DeepSeek-R1-Distill-Qwen-32B	Qwen2-72B
LiveBench Instruction Following	55.7%	—
LMArena Instruction Following	—	1181

Long Context Not comparable

DeepSeek-R1-Distill-Qwen-32B: —, Qwen2-72B: 36.1 (#235)

Long Context benchmarks
Benchmark	DeepSeek-R1-Distill-Qwen-32B	Qwen2-72B
LMArena Longer Query	—	1192

Writing & Preference DeepSeek-R1-Distill-Qwen-32B leads

DeepSeek-R1-Distill-Qwen-32B: 49.6 (#188), Qwen2-72B: 40.8 (#241)

Writing & Preference benchmarks
Benchmark	DeepSeek-R1-Distill-Qwen-32B	Qwen2-72B
LMArena Text	—	1203
LMArena Creative Writing	—	1181
LMArena Multi-Turn	—	1196
LiveBench Language	26.8%	—

Frequently asked questions

Is DeepSeek-R1-Distill-Qwen-32B better than Qwen2-72B?