Noometry

Model comparison

DeepSeek-V3.2-Exp vs Yi-Lightning

DeepSeek-V3.2-Exp is the stronger model overall, scoring 44.9 to 36.5 on the Noometry Index.

Last verified . 18 shared benchmarks.

DeepSeek-V3.2-Exp DeepSeek

44.9

Rank #68 Confirmed

Yi-Lightning 01.AI

36.5

Rank #206 Confirmed

Summary

  • They share 18 benchmarks with published results for both. DeepSeek-V3.2-Exp scores higher in 7 categories and Yi-Lightning in 1 category; 8 gaps are clear of the uncertainty.
  • The widest gap is in coding, where DeepSeek-V3.2-Exp leads 46.4 to 27.0.
  • The biggest single-benchmark swing is Aider Polyglot: 74.2% for DeepSeek-V3.2-Exp and 12.9% for Yi-Lightning.
  • DeepSeek-V3.2-Exp has downloadable open weights; the other is API-only.

Side by side

DeepSeek-V3.2-ExpYi-Lightning
ProviderDeepSeek01.AI
Noometry Index44.936.5
Released2025-09-292024-12-02
WeightsOpenProprietary
Context window164K—
Max output66K—
Input $ / M tokens$0.26—
Output $ / M tokens$0.38—
Results tracked4818

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), Yi-Lightning: 27.0 (#315)

BenchmarkDeepSeek-V3.2-ExpYi-Lightning
Aider Polyglot74.2%12.9%
LMArena Coding14541313
SWE-bench Verified (bash only)70%—
LMArena WebDev1362—
SWE-bench Multilingual59%—
SciCode38.9%—
WeirdML39.5%—

Agentic & Tool Use Not comparable

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

BenchmarkDeepSeek-V3.2-ExpYi-Lightning
Terminal-Bench39.6%—
Berkeley Function Calling Leaderboard56.7%—
TheAgentCompany42.9%—
Vending-Bench 21,034—

Reasoning Yi-Lightning leads

DeepSeek-V3.2-Exp: 22.0 (#185), Yi-Lightning: 24.9 (#137)

BenchmarkDeepSeek-V3.2-ExpYi-Lightning
LMArena Hard Prompts14331302
ARC-AGI-24%—
Kagi LLM Benchmark52.2%—
NYT Connections (extended)36.7%—
ARC-AGI-157%—
CritPt2.9%—
Chess Puzzles14%—
Thematic Generalization65%—
DTBench87.7%—
LMCA29.1%—
Epoch Capabilities Index146.28—

Math DeepSeek-V3.2-Exp leads

DeepSeek-V3.2-Exp: 40.8 (#75), Yi-Lightning: 33.5 (#173)

BenchmarkDeepSeek-V3.2-ExpYi-Lightning
LMArena Math14351300
MathArena Final-Answer Competitions57.7%—
OTIS Mock AIME 2024-202587.8%—
ProofBench8%—
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), Yi-Lightning: 34.8 (#182)

BenchmarkDeepSeek-V3.2-ExpYi-Lightning
LMArena Expert14361286
GPQA Diamond83.4%—
Vectara Hallucination Rate5.3%—

Multilingual DeepSeek-V3.2-Exp leads

DeepSeek-V3.2-Exp: 52.4 (#85), Yi-Lightning: 42.4 (#191)

BenchmarkDeepSeek-V3.2-ExpYi-Lightning
LMArena Non-English14091269
LMArena Chinese14611320
LMArena French14351304
LMArena German14401267
LMArena Japanese13771228
LMArena Korean13721192
LMArena Russian14231255
LMArena Spanish14411315

Instruction Following DeepSeek-V3.2-Exp leads

DeepSeek-V3.2-Exp: 74.7 (#89), Yi-Lightning: 67.6 (#191)

BenchmarkDeepSeek-V3.2-ExpYi-Lightning
LMArena Instruction Following14131278

Long Context DeepSeek-V3.2-Exp leads

DeepSeek-V3.2-Exp: 47.6 (#15), Yi-Lightning: 39.4 (#176)

BenchmarkDeepSeek-V3.2-ExpYi-Lightning
LMArena Longer Query14281297
Fiction.LiveBench83.3%—
CL-bench13.2%—
CL-bench Life9.5%—

Writing & Preference DeepSeek-V3.2-Exp leads

DeepSeek-V3.2-Exp: 62.5 (#73), Yi-Lightning: 50.4 (#178)

BenchmarkDeepSeek-V3.2-ExpYi-Lightning
LMArena Text14251302
LMArena Creative Writing14031280
LMArena Multi-Turn14271311
EQ-Bench Creative Writing1515—

Frequently asked questions

Is DeepSeek-V3.2-Exp better than Yi-Lightning?

DeepSeek-V3.2-Exp is the stronger model overall, scoring 44.9 to 36.5 on the Noometry Index.

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

DeepSeek-V3.2-Exp scores higher on coding benchmarks: 46.4 versus 27.0 in the Noometry coding category.

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

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

Related comparisons

Go deeper