Noometry

Model comparison

Gemini 3.8 Flash vs Qwen2-72B

Gemini 3.8 Flash is the stronger model overall, scoring 61.8 to 30.0 on the Noometry Index.

Last verified . 20 shared benchmarks.

Gemini 3.8 Flash Google

61.8

Rank #11 Confirmed

Qwen2-72B Alibaba (Qwen)

30.0

Rank #300 Confirmed

Summary

  • They share 20 benchmarks with published results for both. Gemini 3.8 Flash scores higher in 9 categories and Qwen2-72B in 0 categories; 9 gaps are clear of the uncertainty.
  • The widest gap is in reasoning, where Gemini 3.8 Flash leads 76.9 to 23.2.
  • The biggest single-benchmark swing is WeirdML: 84.8% for Gemini 3.8 Flash and 11.3% for Qwen2-72B.
  • Qwen2-72B has downloadable open weights; the other is API-only.

Side by side

Gemini 3.8 FlashQwen2-72B
ProviderGoogleAlibaba (Qwen)
Noometry Index61.830.0
Released2026-09-022024-06-07
WeightsProprietaryOpen
Context window1.05M—
Max output66K—
Input $ / M tokens$0.75—
Output $ / M tokens$3.75—
Results tracked5026

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

Category by category

Coding Gemini 3.8 Flash leads

Gemini 3.8 Flash: 59.2 (#15), Qwen2-72B: 29.1 (#310)

BenchmarkGemini 3.8 FlashQwen2-72B
WeirdML84.8%11.3%
LMArena Coding15101196
DeepSWE73.8%—
FrontierCode41.2%—
CursorBench39.6%—
LMArena WebDev1584—
FrontierSWE19.6%—
SciCode56.6%—
BigCodeBench Instruct—38.5%
BigCodeBench Complete—54%
ALE-Bench1,270—

Agentic & Tool Use Gemini 3.8 Flash leads

Gemini 3.8 Flash: 41.8 (#21), Qwen2-72B: 17.0 (#146)

BenchmarkGemini 3.8 FlashQwen2-72B
APEX-Agents64.3%—
Remote Labor Index5.8%—
TheAgentCompany—1.1%
GDP.pdf23.4%—
METR Time Horizons—29.9%
Vending-Bench 25,094—

Reasoning Gemini 3.8 Flash leads

Gemini 3.8 Flash: 76.9 (#5), Qwen2-72B: 23.2 (#181)

BenchmarkGemini 3.8 FlashQwen2-72B
LMArena Hard Prompts15081191
Epoch Capabilities Index156.71125.28
ARC-AGI-289.2%—
NYT Connections (extended)97.4%—
ARC-AGI-198.5%—
CritPt18.3%—
Chess Puzzles61%—
Mystery Game Puzzles47%—
DTBench95.7%—
LMCA52.9%—
Surface Evolver Bench76.9%—

Math Gemini 3.8 Flash leads

Gemini 3.8 Flash: 65.3 (#28), Qwen2-72B: 30.2 (#236)

BenchmarkGemini 3.8 FlashQwen2-72B
LMArena Math15281235
FrontierMath (Tiers 1-3)68.4%—
FrontierMath Tier 422%—
OTIS Mock AIME 2024-202598.9%—
ProofBench48%—
MATH Level 5—39.1%

Knowledge Gemini 3.8 Flash leads

Gemini 3.8 Flash: 74.8 (#2), Qwen2-72B: 21.2 (#275)

BenchmarkGemini 3.8 FlashQwen2-72B
GPQA Diamond95.4%40.8%
LMArena Expert15241171
Humanity's Last Exam44.5%—
SimpleQA Verified69.7%—
MMLU—82.4%

Multimodal Not comparable

Gemini 3.8 Flash: 40.7 (#45), Qwen2-72B: —

BenchmarkGemini 3.8 FlashQwen2-72B
LMArena Vision1314—
Blueprint-Bench 238.6%—
Furniture Assembly31.7%—

Multilingual Gemini 3.8 Flash leads

Gemini 3.8 Flash: 58.0 (#5), Qwen2-72B: 35.9 (#244)

BenchmarkGemini 3.8 FlashQwen2-72B
LMArena Non-English14911176
LMArena Chinese15541240
LMArena French14981170
LMArena German14931151
LMArena Japanese15021111
LMArena Korean14591083
LMArena Russian15151169
LMArena Spanish14851169

Instruction Following Gemini 3.8 Flash leads

Gemini 3.8 Flash: 78.0 (#13), Qwen2-72B: 61.7 (#241)

BenchmarkGemini 3.8 FlashQwen2-72B
LMArena Instruction Following14901181

Long Context Gemini 3.8 Flash leads

Gemini 3.8 Flash: 46.3 (#24), Qwen2-72B: 36.1 (#235)

BenchmarkGemini 3.8 FlashQwen2-72B
LMArena Longer Query15081192

Writing & Preference Gemini 3.8 Flash leads

Gemini 3.8 Flash: 72.2 (#15), Qwen2-72B: 40.8 (#241)

BenchmarkGemini 3.8 FlashQwen2-72B
LMArena Text14991203
LMArena Creative Writing14921181
LMArena Multi-Turn15011196
EQ-Bench Creative Writing1748—

Frequently asked questions

Is Gemini 3.8 Flash better than Qwen2-72B?

Gemini 3.8 Flash is the stronger model overall, scoring 61.8 to 30.0 on the Noometry Index.

Is Gemini 3.8 Flash or Qwen2-72B better for coding?

Gemini 3.8 Flash scores higher on coding benchmarks: 59.2 versus 29.1 in the Noometry coding category.

How many benchmarks do Gemini 3.8 Flash and Qwen2-72B share?

20 benchmarks have published results for both models. Gemini 3.8 Flash has 50 scored results on Noometry and Qwen2-72B has 26.

Related comparisons

Go deeper