Noometry

Model comparison

Gemini 2.0 Flash-Lite vs gpt-oss-20b

Gemini 2.0 Flash-Lite is the stronger model overall, scoring 37.8 to 32.5 on the Noometry Index.

Last verified . 22 shared benchmarks.

Gemini 2.0 Flash-Lite Google

37.8

Rank #194 Confirmed

gpt-oss-20b OpenAI

32.5

Rank #255 Confirmed

Summary

  • They share 22 benchmarks with published results for both. Gemini 2.0 Flash-Lite scores higher in 7 categories and gpt-oss-20b in 1 category; 6 gaps are clear of the uncertainty.
  • The widest gap is in writing & preference, where Gemini 2.0 Flash-Lite leads 51.7 to 35.5.
  • The biggest single-benchmark swing is Omni-MATH: 37.4% for Gemini 2.0 Flash-Lite and 56.5% for gpt-oss-20b.
  • gpt-oss-20b has downloadable open weights; the other is API-only.

Side by side

Gemini 2.0 Flash-Litegpt-oss-20b
ProviderGoogleOpenAI
Noometry Index37.832.5
Released2025-02-052025-08-05
WeightsProprietaryOpen
Context window—131K
Max output—16K
Input $ / M tokens—$0.018
Output $ / M tokens—$0.09
Results tracked3234

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

Category by category

Coding Too close to call

Gemini 2.0 Flash-Lite: 37.9 (#185), gpt-oss-20b: 37.6 (#192)

BenchmarkGemini 2.0 Flash-Litegpt-oss-20b
LMArena Coding13221306
SciCode—34.4%
WeirdML—40.9%
LiveBench Coding47.1%—
ALE-Bench—566.05

Agentic & Tool Use Not comparable

Gemini 2.0 Flash-Lite: —, gpt-oss-20b: 9.3 (#154)

BenchmarkGemini 2.0 Flash-Litegpt-oss-20b
Terminal-Bench—3.4%

Reasoning Gemini 2.0 Flash-Lite leads

Gemini 2.0 Flash-Lite: 22.0 (#210), gpt-oss-20b: 19.3 (#261)

BenchmarkGemini 2.0 Flash-Litegpt-oss-20b
LMArena Hard Prompts13241274
DTBench52.5%68%
Kagi LLM Benchmark—53.2%
CritPt—1.4%
Chess Puzzles—4%
LiveBench Reasoning50.1%—
LiveBench Data Analysis65.5%—
LMCA—14.5%
Epoch Capabilities Index—137.82
ForecastBench57.1—
LiveBench54.3%—

Math gpt-oss-20b leads

Gemini 2.0 Flash-Lite: 34.1 (#196), gpt-oss-20b: 39.4 (#103)

BenchmarkGemini 2.0 Flash-Litegpt-oss-20b
Omni-MATH37.4%56.5%
LMArena Math13091317
OTIS Mock AIME 2024-2025—65.3%
LiveBench Math58.1%—

Knowledge Too close to call

Gemini 2.0 Flash-Lite: 35.0 (#189), gpt-oss-20b: 34.6 (#195)

BenchmarkGemini 2.0 Flash-Litegpt-oss-20b
MMLU-Pro72%74%
GPQA (HELM)50%59.4%
LMArena Expert13051258
GPQA Diamond—60.8%

Multimodal Not comparable

Gemini 2.0 Flash-Lite: 31.2 (#109), gpt-oss-20b: —

BenchmarkGemini 2.0 Flash-Litegpt-oss-20b
LMArena Vision1100—

Multilingual Gemini 2.0 Flash-Lite leads

Gemini 2.0 Flash-Lite: 46.0 (#161), gpt-oss-20b: 42.2 (#197)

BenchmarkGemini 2.0 Flash-Litegpt-oss-20b
LMArena Non-English13231268
LMArena Chinese13391314
LMArena German13061255
LMArena Japanese13011244
LMArena Korean13251236
LMArena Russian13281278
LMArena Spanish13131267
LMArena French1347—

Instruction Following Gemini 2.0 Flash-Lite leads

Gemini 2.0 Flash-Lite: 70.4 (#163), gpt-oss-20b: 61.8 (#240)

BenchmarkGemini 2.0 Flash-Litegpt-oss-20b
IFEval82.4%73.2%
LMArena Instruction Following13051236
LiveBench Instruction Following78.3%—

Long Context Gemini 2.0 Flash-Lite leads

Gemini 2.0 Flash-Lite: 40.1 (#160), gpt-oss-20b: 37.9 (#209)

BenchmarkGemini 2.0 Flash-Litegpt-oss-20b
LMArena Longer Query13201250

Writing & Preference Gemini 2.0 Flash-Lite leads

Gemini 2.0 Flash-Lite: 51.7 (#177), gpt-oss-20b: 35.5 (#265)

BenchmarkGemini 2.0 Flash-Litegpt-oss-20b
LMArena Text13301287
LMArena Creative Writing13191201
WildBench79%73.7%
LMArena Multi-Turn13071268
EQ-Bench Creative Writing—666
LiveBench Language34.3%—

Frequently asked questions

Is Gemini 2.0 Flash-Lite better than gpt-oss-20b?

Gemini 2.0 Flash-Lite is the stronger model overall, scoring 37.8 to 32.5 on the Noometry Index.

Is Gemini 2.0 Flash-Lite or gpt-oss-20b better for coding?

They score almost the same on coding (37.9 vs 37.6); test both on your own repository before choosing.

How many benchmarks do Gemini 2.0 Flash-Lite and gpt-oss-20b share?

22 benchmarks have published results for both models. Gemini 2.0 Flash-Lite has 32 scored results on Noometry and gpt-oss-20b has 34.

Related comparisons

Go deeper