Model comparison

o1 vs Qwen2.5-Coder-32B

o1 is the stronger model overall, scoring 40.9 to 33.4 on the Noometry Index. Qwen2.5-Coder-32B costs 35× less per token, which makes it the better buy when o1's lead doesn't matter for your workload.

Last verified October 11, 2026. 23 shared benchmarks.

o1 OpenAI

40.9

Rank #143 Confirmed

Qwen2.5-Coder-32B

Alibaba (Qwen)

33.4

Rank #245 Confirmed

Summary

They share 23 benchmarks with published results for both. o1 scores higher in 8 categories and Qwen2.5-Coder-32B in 0 categories; 8 gaps are clear of the uncertainty.
The widest gap is in coding, where o1 leads 46.1 to 22.6.
The biggest single-benchmark swing is LiveBench Reasoning: 91.6% for o1 and 42.1% for Qwen2.5-Coder-32B.
Qwen2.5-Coder-32B is cheaper at $0.66 / $1 per million input/output tokens, against $15 / $60 for o1.
o1 accepts more context: 200K tokens versus 33K.
Qwen2.5-Coder-32B has downloadable open weights; the other is API-only.

Side by side

o1 and Qwen2.5-Coder-32B specifications
	o1	Qwen2.5-Coder-32B
Provider	OpenAI	Alibaba (Qwen)
Noometry Index	40.9	33.4
Released	2024-09-12	2024-09-18
Weights	Proprietary	Open
Context window	200K	33K
Max output	100K	29K
Input $ / M tokens	$15	$0.66
Output $ / M tokens	$60	$1
Results tracked	52	31

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

Category by category

Coding o1 leads

o1: 46.1 (#70), Qwen2.5-Coder-32B: 22.6 (#333)

Coding benchmarks
Benchmark	o1	Qwen2.5-Coder-32B
Aider Polyglot	61.7%	16.4%
LiveBench Coding	69.7%	56.9%
LMArena Coding	1367	1276
HumanEval+	89%	87.2%
MBPP+	80.2%	77%
SWE-bench Verified (bash only)	—	9%
WeirdML	47.6%	—
BigCodeBench Instruct	—	49%
BigCodeBench Complete	—	58%
CadEval	56%	—

Agentic & Tool Use Not comparable

o1: 24.6 (#117), Qwen2.5-Coder-32B: —

Agentic & Tool Use benchmarks
Benchmark	o1	Qwen2.5-Coder-32B
Cybench	10%	—
METR Time Horizons	51.1%	—

Reasoning o1 leads

o1: 27.9 (#111), Qwen2.5-Coder-32B: 21.2 (#225)

Reasoning benchmarks
Benchmark	o1	Qwen2.5-Coder-32B
LiveBench Reasoning	91.6%	42.1%
LMArena Hard Prompts	1371	1251
LiveBench Data Analysis	65.5%	49.9%
Epoch Capabilities Index	141.91	119.49
LiveBench	75.7%	46.2%
SimpleBench	41.7%	—
ARC-AGI-1	30.7%	—
Chess Puzzles	15%	—
EnigmaEval	5.7%	—
DTBench	74.7%	—
LMCA	22.3%	—
HellaSwag	—	83%
WinoGrande	—	80.8%

Math o1 leads

o1: 36.1 (#175), Qwen2.5-Coder-32B: 33.3 (#204)

Math benchmarks
Benchmark	o1	Qwen2.5-Coder-32B
LiveBench Math	80.3%	46.6%
LMArena Math	1388	1251
FrontierMath (Tiers 1-3)	14.7%	—
OTIS Mock AIME 2024-2025	73.3%	—
MATH Level 5	94.7%	—
FrontierMath (Feb 2025 set)	9.3%	—
GSM8K	—	93%

Knowledge o1 leads

o1: 41.5 (#110), Qwen2.5-Coder-32B: 33.4 (#203)

Knowledge benchmarks
Benchmark	o1	Qwen2.5-Coder-32B
LMArena Expert	1361	1221
GPQA Diamond	76.8%	—
Humanity's Last Exam	8%	—
SimpleQA Verified	41.1%	—
Confabulations	11.7%	—
ARC (AI2) Challenge	—	70.5%
MMLU	—	79.1%

Multimodal Not comparable

o1: 34.2 (#93), Qwen2.5-Coder-32B: —

Multimodal benchmarks
Benchmark	o1	Qwen2.5-Coder-32B
LMArena Vision	1168	—
GeoBench	80%	—
VPCT	37%	—
SpatialViz-Bench	41.4%	—

Multilingual o1 leads

o1: 48.6 (#142), Qwen2.5-Coder-32B: 37.8 (#235)

Multilingual benchmarks
Benchmark	o1	Qwen2.5-Coder-32B
LMArena Non-English	1358	1205
LMArena Chinese	1394	1222
LMArena Russian	1356	1228
LMArena French	1344	—
LMArena German	1337	—
LMArena Japanese	1346	—
LMArena Korean	1396	—
LMArena Spanish	1345	—

Instruction Following o1 leads

o1: 74.8 (#86), Qwen2.5-Coder-32B: 61.4 (#245)

Instruction Following benchmarks
Benchmark	o1	Qwen2.5-Coder-32B
LiveBench Instruction Following	81.5%	58.7%
LMArena Instruction Following	1367	1223

Long Context o1 leads

o1: 50.3 (#9), Qwen2.5-Coder-32B: 38.0 (#208)

Long Context benchmarks
Benchmark	o1	Qwen2.5-Coder-32B
LMArena Longer Query	1378	1251
Fiction.LiveBench	83.3%	—

Writing & Preference o1 leads

o1: 55.6 (#144), Qwen2.5-Coder-32B: 41.6 (#240)

Writing & Preference benchmarks
Benchmark	o1	Qwen2.5-Coder-32B
LMArena Text	1366	1230
LMArena Creative Writing	1348	1174
LMArena Multi-Turn	1369	1222
LiveBench Language	65.4%	23.3%
Short-Story Creative Writing	70.2%	—

Frequently asked questions

Is o1 better than Qwen2.5-Coder-32B?

Which is cheaper, o1 or Qwen2.5-Coder-32B?

Qwen2.5-Coder-32B is cheaper. It lists at $0.66 per million input tokens and $1 per million output tokens; o1 lists at $15 and $60.

Is o1 or Qwen2.5-Coder-32B better for coding?

o1 scores higher on coding benchmarks: 46.1 versus 22.6 in the Noometry coding category.

Which has the bigger context window?

o1 does, with 200K tokens against 33K.

How many benchmarks do o1 and Qwen2.5-Coder-32B share?

23 benchmarks have published results for both models. o1 has 52 scored results on Noometry and Qwen2.5-Coder-32B has 31.