Model comparison

DeepSeek-V3 vs Qwen2.5-Coder (1.5B)

DeepSeek-V3 has enough public results to be ranked (#166); Qwen2.5-Coder (1.5B) does not yet, so treat this comparison as directional.

Last verified October 10, 2026. 5 shared benchmarks.

DeepSeek-V3

DeepSeek

39.5

Rank #166 Confirmed

Qwen2.5-Coder (1.5B)

Alibaba (Qwen)

—

Unranked

Summary

They share 5 benchmarks with published results for both.

Side by side

DeepSeek-V3 and Qwen2.5-Coder (1.5B) specifications
	DeepSeek-V3	Qwen2.5-Coder (1.5B)
Provider	DeepSeek	Alibaba (Qwen)
Noometry Index	39.5	—
Released	2024-12-26	2024-09-18
Weights	Open	Open
Context window	164K	—
Max output	164K	—
Input $ / M tokens	$0.24	—
Output $ / M tokens	$0.90	—
Results tracked	60	6

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

Category by category

Coding Not comparable

DeepSeek-V3: 42.3 (#106), Qwen2.5-Coder (1.5B): —

Coding benchmarks
Benchmark	DeepSeek-V3	Qwen2.5-Coder (1.5B)
Aider Polyglot	55.1%	—
SciCode	35.8%	—
WeirdML	36.1%	—
BigCodeBench Instruct	50%	—
LiveBench Coding	70.9%	—
LMArena Coding	1368	—
BigCodeBench Complete	62.2%	—
HumanEval+	86.6%	—
MBPP+	73%	—

Agentic & Tool Use Not comparable

DeepSeek-V3: —, Qwen2.5-Coder (1.5B): —

Agentic & Tool Use benchmarks
Benchmark	DeepSeek-V3	Qwen2.5-Coder (1.5B)
METR Time Horizons	49.6%	—

Reasoning Not comparable

DeepSeek-V3: 20.5 (#236), Qwen2.5-Coder (1.5B): —

Reasoning benchmarks
Benchmark	DeepSeek-V3	Qwen2.5-Coder (1.5B)
Epoch Capabilities Index	135.94	113.14
HellaSwag	88.9%	76.8%
WinoGrande	85.2%	72.9%
SimpleBench	27.2%	—
Kagi LLM Benchmark	52.3%	—
CritPt	0%	—
LiveBench Reasoning	65.8%	—
LMArena Hard Prompts	1365	—
DTBench	64.8%	—
LiveBench Data Analysis	60.9%	—
LMCA	15.5%	—
BIG-Bench Hard	87.5%	—
ForecastBench	59.1	—
LiveBench	66.9%	—
PIQA	84.7%	—

Math Not comparable

DeepSeek-V3: 32.1 (#219), Qwen2.5-Coder (1.5B): —

Math benchmarks
Benchmark	DeepSeek-V3	Qwen2.5-Coder (1.5B)
OTIS Mock AIME 2024-2025	37.8%	—
Omni-MATH	40.3%	—
LiveBench Math	73.5%	—
LMArena Math	1373	—
MATH Level 5	75.5%	—
FrontierMath (Feb 2025 set)	1.7%	—
GSM8K	—	86.7%

Knowledge Not comparable

DeepSeek-V3: 37.5 (#155), Qwen2.5-Coder (1.5B): —

Knowledge benchmarks
Benchmark	DeepSeek-V3	Qwen2.5-Coder (1.5B)
ARC (AI2) Challenge	95.3%	60.9%
MMLU	87.2%	68%
GPQA Diamond	67.6%	—
MMLU-Pro	72.3%	—
Confabulations	26.1%	—
Vectara Hallucination Rate	6.1%	—
GPQA (HELM)	53.8%	—
LMArena Expert	1351	—
TriviaQA	82.9%	—

Multilingual Not comparable

DeepSeek-V3: 48.5 (#143), Qwen2.5-Coder (1.5B): —

Multilingual benchmarks
Benchmark	DeepSeek-V3	Qwen2.5-Coder (1.5B)
LMArena Non-English	1358	—
LMArena Chinese	1391	—
LMArena French	1385	—
LMArena German	1374	—
LMArena Japanese	1333	—
LMArena Korean	1319	—
LMArena Russian	1373	—
LMArena Spanish	1358	—

Instruction Following Not comparable

DeepSeek-V3: 72.8 (#130), Qwen2.5-Coder (1.5B): —

Instruction Following benchmarks
Benchmark	DeepSeek-V3	Qwen2.5-Coder (1.5B)
LiveBench Instruction Following	81.5%	—
IFEval	83.2%	—
LMArena Instruction Following	1345	—

Long Context Not comparable

DeepSeek-V3: 34.0 (#253), Qwen2.5-Coder (1.5B): —

Long Context benchmarks
Benchmark	DeepSeek-V3	Qwen2.5-Coder (1.5B)
Fiction.LiveBench	50%	—
LMArena Longer Query	1352	—

Writing & Preference Not comparable

DeepSeek-V3: 57.4 (#130), Qwen2.5-Coder (1.5B): —

Writing & Preference benchmarks
Benchmark	DeepSeek-V3	Qwen2.5-Coder (1.5B)
LMArena Text	1375	—
LMArena Creative Writing	1364	—
Short-Story Creative Writing	77%	—
EQ-Bench Creative Writing	1472	—
WildBench	83%	—
LMArena Multi-Turn	1389	—
LiveBench Language	49.1%	—

Frequently asked questions

Is DeepSeek-V3 better than Qwen2.5-Coder (1.5B)?