Noometry

Model comparison

DeepSeek-V3.1-Terminus vs Gemini 2.0 Flash-Lite

DeepSeek-V3.1-Terminus is the stronger model overall, scoring 42.6 to 37.3 on the Noometry Index.

Last verified . 11 shared benchmarks.

DeepSeek-V3.1-Terminus DeepSeek

42.6

Rank #96 Confirmed

Gemini 2.0 Flash-Lite Google

37.3

Rank #189 Confirmed

Summary

  • They share 11 benchmarks with published results for both. DeepSeek-V3.1-Terminus scores higher in 7 categories and Gemini 2.0 Flash-Lite in 0 categories; 7 gaps are clear of the uncertainty.
  • The widest gap is in writing & preference, where DeepSeek-V3.1-Terminus leads 61.3 to 51.9.
  • The biggest single-benchmark swing is DTBench: 81.3% for DeepSeek-V3.1-Terminus and 52.5% for Gemini 2.0 Flash-Lite.
  • DeepSeek-V3.1-Terminus has downloadable open weights; the other is API-only.

Side by side

DeepSeek-V3.1-TerminusGemini 2.0 Flash-Lite
ProviderDeepSeekGoogle
Noometry Index42.637.3
Released2025-09-222025-02-05
WeightsOpenProprietary
Context window164K—
Max output66K—
Input $ / M tokens$0.30—
Output $ / M tokens$1—
Results tracked1632

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

Category by category

Coding DeepSeek-V3.1-Terminus leads

DeepSeek-V3.1-Terminus: 41.5 (#114), Gemini 2.0 Flash-Lite: 37.4 (#181)

BenchmarkDeepSeek-V3.1-TerminusGemini 2.0 Flash-Lite
LMArena Coding14261322
SciCode40.6%—
LiveBench Coding—47.1%
ALE-Bench745.17—

Reasoning DeepSeek-V3.1-Terminus leads

DeepSeek-V3.1-Terminus: 25.7 (#126), Gemini 2.0 Flash-Lite: 21.2 (#203)

BenchmarkDeepSeek-V3.1-TerminusGemini 2.0 Flash-Lite
LMArena Hard Prompts14271324
DTBench81.3%52.5%
Kagi LLM Benchmark57.4%—
CritPt1.7%—
LiveBench Reasoning—50.1%
LiveBench Data Analysis—65.5%
LMCA28.6%—
ForecastBench—57.1
LiveBench—54.3%

Math DeepSeek-V3.1-Terminus leads

DeepSeek-V3.1-Terminus: 35.7 (#138), Gemini 2.0 Flash-Lite: 31.8 (#192)

BenchmarkDeepSeek-V3.1-TerminusGemini 2.0 Flash-Lite
LMArena Math14021309
Omni-MATH—37.4%
LiveBench Math—58.1%

Knowledge Not comparable

DeepSeek-V3.1-Terminus: —, Gemini 2.0 Flash-Lite: 34.6 (#185)

BenchmarkDeepSeek-V3.1-TerminusGemini 2.0 Flash-Lite
MMLU-Pro—72%
GPQA (HELM)—50%
LMArena Expert—1305

Multimodal Not comparable

DeepSeek-V3.1-Terminus: —, Gemini 2.0 Flash-Lite: 30.6 (#103)

BenchmarkDeepSeek-V3.1-TerminusGemini 2.0 Flash-Lite
LMArena Vision—1100

Multilingual DeepSeek-V3.1-Terminus leads

DeepSeek-V3.1-Terminus: 52.3 (#87), Gemini 2.0 Flash-Lite: 46.2 (#156)

BenchmarkDeepSeek-V3.1-TerminusGemini 2.0 Flash-Lite
LMArena Non-English14071323
LMArena Russian14341328
LMArena Chinese—1339
LMArena French—1346
LMArena German—1306
LMArena Japanese—1302
LMArena Korean—1325
LMArena Spanish—1313

Instruction Following DeepSeek-V3.1-Terminus leads

DeepSeek-V3.1-Terminus: 74.4 (#96), Gemini 2.0 Flash-Lite: 70.6 (#158)

BenchmarkDeepSeek-V3.1-TerminusGemini 2.0 Flash-Lite
LMArena Instruction Following14061305
LiveBench Instruction Following—78.3%
IFEval—82.4%

Long Context DeepSeek-V3.1-Terminus leads

DeepSeek-V3.1-Terminus: 43.5 (#88), Gemini 2.0 Flash-Lite: 40.1 (#155)

BenchmarkDeepSeek-V3.1-TerminusGemini 2.0 Flash-Lite
LMArena Longer Query14261321

Writing & Preference DeepSeek-V3.1-Terminus leads

DeepSeek-V3.1-Terminus: 61.3 (#87), Gemini 2.0 Flash-Lite: 51.9 (#172)

BenchmarkDeepSeek-V3.1-TerminusGemini 2.0 Flash-Lite
LMArena Text14201330
LMArena Creative Writing14041319
LMArena Multi-Turn14111307
WildBench—79%
LiveBench Language—34.3%

Frequently asked questions

Is DeepSeek-V3.1-Terminus better than Gemini 2.0 Flash-Lite?

DeepSeek-V3.1-Terminus is the stronger model overall, scoring 42.6 to 37.3 on the Noometry Index.

Is DeepSeek-V3.1-Terminus or Gemini 2.0 Flash-Lite better for coding?

DeepSeek-V3.1-Terminus scores higher on coding benchmarks: 41.5 versus 37.4 in the Noometry coding category.

How many benchmarks do DeepSeek-V3.1-Terminus and Gemini 2.0 Flash-Lite share?

11 benchmarks have published results for both models. DeepSeek-V3.1-Terminus has 16 scored results on Noometry and Gemini 2.0 Flash-Lite has 32.

Related comparisons

Go deeper