Model comparison

DeepSeek-R1 vs Qwen3-4B

DeepSeek-R1 is the stronger model overall, scoring 42.3 to 31.9 on the Noometry Index.

Last verified October 10, 2026. 3 shared benchmarks.

DeepSeek-R1

DeepSeek

42.3

Rank #115 Confirmed

Qwen3-4B

Alibaba (Qwen)

31.9

Rank #264 Confirmed

Summary

They share 3 benchmarks with published results for both. DeepSeek-R1 scores higher in 3 categories and Qwen3-4B in 1 category; 3 gaps are clear of the uncertainty.
The widest gap is in math, where DeepSeek-R1 leads 43.8 to 29.7.
The biggest single-benchmark swing is GPQA Diamond: 76.3% for DeepSeek-R1 and 52.3% for Qwen3-4B.
Qwen3-4B has downloadable open weights; the other is API-only.

Side by side

DeepSeek-R1 and Qwen3-4B specifications
	DeepSeek-R1	Qwen3-4B
Provider	DeepSeek	Alibaba (Qwen)
Noometry Index	42.3	31.9
Released	2025-01-20	2025-04-29
Weights	Proprietary	Open
Context window	164K	—
Max output	64K	—
Input $ / M tokens	$0.50	—
Output $ / M tokens	$2.15	—
Results tracked	52	6

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

Category by category

Coding Not comparable

DeepSeek-R1: 46.3 (#68), Qwen3-4B: —

Coding benchmarks
Benchmark	DeepSeek-R1	Qwen3-4B
Aider Polyglot	71.4%	—
SciCode	35.7%	—
WeirdML	41.6%	—
LiveBench Coding	66.7%	—
LMArena Coding	1427	—
ALE-Bench	804.12	—
AlgoTune	1.7	—

Agentic & Tool Use DeepSeek-R1 leads

DeepSeek-R1: 30.7 (#75), Qwen3-4B: 27.6 (#100)

Agentic & Tool Use benchmarks
Benchmark	DeepSeek-R1	Qwen3-4B
Berkeley Function Calling Leaderboard	—	35.7%
DeepResearch Bench	35.1%	—
BALROG	34.9%	—
METR Time Horizons	53.8%	—

Reasoning Too close to call

DeepSeek-R1: 18.6 (#278), Qwen3-4B: 19.2 (#268)

Reasoning benchmarks
Benchmark	DeepSeek-R1	Qwen3-4B
ARC-AGI-2	1.3%	—
SimpleBench	40.8%	—
Kagi LLM Benchmark	69.4%	—
ARC-AGI-1	21.2%	—
CritPt	1.1%	—
Chess Puzzles	—	4%
LiveBench Reasoning	83.2%	—
LMArena Hard Prompts	1416	—
LiveBench Data Analysis	69.8%	—
Epoch Capabilities Index	141.29	—
ForecastBench	60	—
LiveBench	71.6%	—

Math DeepSeek-R1 leads

DeepSeek-R1: 43.8 (#79), Qwen3-4B: 29.7 (#240)

Math benchmarks
Benchmark	DeepSeek-R1	Qwen3-4B
OTIS Mock AIME 2024-2025	66.4%	52.2%
MathArena Final-Answer Competitions	—	38.5%
Omni-MATH	42.4%	—
LiveBench Math	80.7%	—
LMArena Math	1400	—
MATH Level 5	96.6%	—

Knowledge DeepSeek-R1 leads

DeepSeek-R1: 44.5 (#87), Qwen3-4B: 33.0 (#208)

Knowledge benchmarks
Benchmark	DeepSeek-R1	Qwen3-4B
GPQA Diamond	76.3%	52.3%
Vectara Hallucination Rate	11.3%	5.7%
MMLU-Pro	79.3%	—
Confabulations	12.7%	—
GPQA (HELM)	66.6%	—
LMArena Expert	1394	—

Multilingual Not comparable

DeepSeek-R1: 52.4 (#85), Qwen3-4B: —

Multilingual benchmarks
Benchmark	DeepSeek-R1	Qwen3-4B
LMArena Non-English	1412	—
LMArena Chinese	1442	—
LMArena French	1417	—
LMArena German	1404	—
LMArena Japanese	1391	—
LMArena Korean	1360	—
LMArena Russian	1423	—
LMArena Spanish	1411	—

Instruction Following Not comparable

DeepSeek-R1: 72.0 (#143), Qwen3-4B: —

Instruction Following benchmarks
Benchmark	DeepSeek-R1	Qwen3-4B
LiveBench Instruction Following	80.5%	—
IFEval	78.4%	—
LMArena Instruction Following	1382	—

Long Context Not comparable

DeepSeek-R1: 45.4 (#36), Qwen3-4B: —

Long Context benchmarks
Benchmark	DeepSeek-R1	Qwen3-4B
Fiction.LiveBench	75%	—
LMArena Longer Query	1391	—

Writing & Preference Not comparable

DeepSeek-R1: 61.4 (#88), Qwen3-4B: —

Writing & Preference benchmarks
Benchmark	DeepSeek-R1	Qwen3-4B
LMArena Text	1428	—
LMArena Creative Writing	1405	—
Short-Story Creative Writing	83%	—
EQ-Bench Creative Writing	1500	—
WildBench	82.8%	—
LMArena Multi-Turn	1405	—
LiveBench Language	48.5%	—

Frequently asked questions

Is DeepSeek-R1 better than Qwen3-4B?