Model comparison

DeepSeek-V3.2-Exp vs Nemotron 3.5 Lightning

DeepSeek-V3.2-Exp is the stronger model overall, scoring 44.9 to 39.3 on the Noometry Index. Nemotron 3.5 Lightning costs 3.4× less per token, which makes it the better buy when DeepSeek-V3.2-Exp's lead doesn't matter for your workload.

Last verified October 10, 2026. 18 shared benchmarks.

DeepSeek-V3.2-Exp

DeepSeek

44.9

Rank #68 Confirmed

Nemotron 3.5 Lightning

NVIDIA

39.3

Rank #156 Confirmed

Summary

They share 18 benchmarks with published results for both. DeepSeek-V3.2-Exp scores higher in 7 categories and Nemotron 3.5 Lightning in 1 category; 8 gaps are clear of the uncertainty.
The widest gap is in knowledge, where DeepSeek-V3.2-Exp leads 51.6 to 36.9.
Nemotron 3.5 Lightning is cheaper at $0.06 / $0.16 per million input/output tokens, against $0.26 / $0.38 for DeepSeek-V3.2-Exp.
Nemotron 3.5 Lightning accepts more context: 262K tokens versus 164K.

Side by side

DeepSeek-V3.2-Exp and Nemotron 3.5 Lightning specifications
	DeepSeek-V3.2-Exp	Nemotron 3.5 Lightning
Provider	DeepSeek	NVIDIA
Noometry Index	44.9	39.3
Released	2025-09-29	2026-08-11
Weights	Open	Open
Context window	164K	262K
Max output	66K	262K
Input $ / M tokens	$0.26	$0.06
Output $ / M tokens	$0.38	$0.16
Results tracked	48	18

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

Category by category

Coding DeepSeek-V3.2-Exp leads

DeepSeek-V3.2-Exp: 46.4 (#61), Nemotron 3.5 Lightning: 39.6 (#143)

Coding benchmarks
Benchmark	DeepSeek-V3.2-Exp	Nemotron 3.5 Lightning
LMArena Coding	1454	1370
SWE-bench Verified (bash only)	70%	—
Aider Polyglot	74.2%	—
LMArena WebDev	1362	—
SWE-bench Multilingual	59%	—
SciCode	38.9%	—
WeirdML	39.5%	—

Agentic & Tool Use Not comparable

DeepSeek-V3.2-Exp: 37.9 (#31), Nemotron 3.5 Lightning: —

Agentic & Tool Use benchmarks
Benchmark	DeepSeek-V3.2-Exp	Nemotron 3.5 Lightning
Terminal-Bench	39.6%	—
Berkeley Function Calling Leaderboard	56.7%	—
TheAgentCompany	42.9%	—
Vending-Bench 2	1,034	—

Reasoning Nemotron 3.5 Lightning leads

DeepSeek-V3.2-Exp: 22.0 (#185), Nemotron 3.5 Lightning: 25.7 (#128)

Reasoning benchmarks
Benchmark	DeepSeek-V3.2-Exp	Nemotron 3.5 Lightning
LMArena Hard Prompts	1433	1335
ARC-AGI-2	4%	—
Kagi LLM Benchmark	52.2%	—
NYT Connections (extended)	36.7%	—
ARC-AGI-1	57%	—
CritPt	2.9%	—
Chess Puzzles	14%	—
Thematic Generalization	65%	—
DTBench	87.7%	—
LMCA	29.1%	—
Epoch Capabilities Index	146.28	—

Math DeepSeek-V3.2-Exp leads

DeepSeek-V3.2-Exp: 40.8 (#75), Nemotron 3.5 Lightning: 34.7 (#156)

Math benchmarks
Benchmark	DeepSeek-V3.2-Exp	Nemotron 3.5 Lightning
LMArena Math	1435	1355
MathArena Final-Answer Competitions	57.7%	—
OTIS Mock AIME 2024-2025	87.8%	—
ProofBench	8%	—
FrontierMath (Feb 2025 set)	22.1%	—
FrontierMath Tier 4 (v1)	2.1%	—

Knowledge DeepSeek-V3.2-Exp leads

DeepSeek-V3.2-Exp: 51.6 (#61), Nemotron 3.5 Lightning: 36.9 (#154)

Knowledge benchmarks
Benchmark	DeepSeek-V3.2-Exp	Nemotron 3.5 Lightning
LMArena Expert	1436	1354
GPQA Diamond	83.4%	—
Vectara Hallucination Rate	5.3%	—

Multilingual DeepSeek-V3.2-Exp leads

DeepSeek-V3.2-Exp: 52.4 (#85), Nemotron 3.5 Lightning: 44.0 (#175)

Multilingual benchmarks
Benchmark	DeepSeek-V3.2-Exp	Nemotron 3.5 Lightning
LMArena Non-English	1409	1292
LMArena Chinese	1461	1356
LMArena French	1435	1369
LMArena German	1440	1280
LMArena Japanese	1377	1213
LMArena Korean	1372	1233
LMArena Russian	1423	1257
LMArena Spanish	1441	1321

Instruction Following DeepSeek-V3.2-Exp leads

DeepSeek-V3.2-Exp: 74.7 (#89), Nemotron 3.5 Lightning: 69.8 (#166)

Instruction Following benchmarks
Benchmark	DeepSeek-V3.2-Exp	Nemotron 3.5 Lightning
LMArena Instruction Following	1413	1317

Long Context DeepSeek-V3.2-Exp leads

DeepSeek-V3.2-Exp: 47.6 (#15), Nemotron 3.5 Lightning: 39.9 (#160)

Long Context benchmarks
Benchmark	DeepSeek-V3.2-Exp	Nemotron 3.5 Lightning
LMArena Longer Query	1428	1312
Fiction.LiveBench	83.3%	—
CL-bench	13.2%	—
CL-bench Life	9.5%	—

Writing & Preference DeepSeek-V3.2-Exp leads

DeepSeek-V3.2-Exp: 62.5 (#73), Nemotron 3.5 Lightning: 48.5 (#198)

Writing & Preference benchmarks
Benchmark	DeepSeek-V3.2-Exp	Nemotron 3.5 Lightning
LMArena Text	1425	1324
LMArena Creative Writing	1403	1254
EQ-Bench Creative Writing	1515	1280
LMArena Multi-Turn	1427	1327

Frequently asked questions

Is DeepSeek-V3.2-Exp better than Nemotron 3.5 Lightning?

Which is cheaper, DeepSeek-V3.2-Exp or Nemotron 3.5 Lightning?

Nemotron 3.5 Lightning is cheaper. It lists at $0.06 per million input tokens and $0.16 per million output tokens; DeepSeek-V3.2-Exp lists at $0.26 and $0.38.

Is DeepSeek-V3.2-Exp or Nemotron 3.5 Lightning better for coding?

DeepSeek-V3.2-Exp scores higher on coding benchmarks: 46.4 versus 39.6 in the Noometry coding category.

Which has the bigger context window?

Nemotron 3.5 Lightning does, with 262K tokens against 164K.

How many benchmarks do DeepSeek-V3.2-Exp and Nemotron 3.5 Lightning share?

18 benchmarks have published results for both models. DeepSeek-V3.2-Exp has 48 scored results on Noometry and Nemotron 3.5 Lightning has 18.