Model comparison

Gemini 3.8 Flash vs Phi-3.5-MoE

Gemini 3.8 Flash has enough public results to be ranked (#11); Phi-3.5-MoE does not yet, so treat this comparison as directional.

Last verified . 0 shared benchmarks.

Gemini 3.8 Flash Google

61.8

Rank #11 Confirmed

Phi-3.5-MoE Microsoft

—

Unranked

Summary

  • Phi-3.5-MoE has downloadable open weights; the other is API-only.

Side by side

Gemini 3.8 Flash and Phi-3.5-MoE specifications
Gemini 3.8 FlashPhi-3.5-MoE
ProviderGoogleMicrosoft
Noometry Index61.8—
Released2026-09-022024-08-17
WeightsProprietaryOpen
Context window1.05M—
Max output66K—
Input $ / M tokens$0.75—
Output $ / M tokens$3.75—
Results tracked503

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

Category by category

Coding Not comparable

Gemini 3.8 Flash: 59.2 (#15), Phi-3.5-MoE: —

Coding benchmarks
BenchmarkGemini 3.8 FlashPhi-3.5-MoE
DeepSWE73.8%—
FrontierCode41.2%—
CursorBench39.6%—
LMArena WebDev1584—
FrontierSWE19.6%—
SciCode56.6%—
WeirdML84.8%—
LMArena Coding1510—
ALE-Bench1,270—

Agentic & Tool Use Not comparable

Gemini 3.8 Flash: 41.8 (#21), Phi-3.5-MoE: —

Agentic & Tool Use benchmarks
BenchmarkGemini 3.8 FlashPhi-3.5-MoE
APEX-Agents64.3%—
Remote Labor Index5.8%—
GDP.pdf23.4%—
Vending-Bench 25,094—

Reasoning Not comparable

Gemini 3.8 Flash: 76.9 (#5), Phi-3.5-MoE: —

Reasoning benchmarks
BenchmarkGemini 3.8 FlashPhi-3.5-MoE
ARC-AGI-289.2%—
NYT Connections (extended)97.4%—
ARC-AGI-198.5%—
CritPt18.3%—
Chess Puzzles61%—
LMArena Hard Prompts1508—
Mystery Game Puzzles47%—
DTBench95.7%—
LMCA52.9%—
Surface Evolver Bench76.9%—
Epoch Capabilities Index156.71—
PIQA—88.6%

Math Not comparable

Gemini 3.8 Flash: 65.3 (#28), Phi-3.5-MoE: —

Math benchmarks
BenchmarkGemini 3.8 FlashPhi-3.5-MoE
FrontierMath (Tiers 1-3)68.4%—
FrontierMath Tier 422%—
OTIS Mock AIME 2024-202598.9%—
ProofBench48%—
LMArena Math1528—
GSM8K—88.7%

Knowledge Not comparable

Gemini 3.8 Flash: 74.8 (#2), Phi-3.5-MoE: —

Knowledge benchmarks
BenchmarkGemini 3.8 FlashPhi-3.5-MoE
GPQA Diamond95.4%—
Humanity's Last Exam44.5%—
SimpleQA Verified69.7%—
LMArena Expert1524—
BoolQ—84.6%

Multimodal Not comparable

Gemini 3.8 Flash: 40.7 (#45), Phi-3.5-MoE: —

Multimodal benchmarks
BenchmarkGemini 3.8 FlashPhi-3.5-MoE
LMArena Vision1314—
Blueprint-Bench 238.6%—
Furniture Assembly31.7%—

Multilingual Not comparable

Gemini 3.8 Flash: 58.0 (#5), Phi-3.5-MoE: —

Multilingual benchmarks
BenchmarkGemini 3.8 FlashPhi-3.5-MoE
LMArena Non-English1491—
LMArena Chinese1554—
LMArena French1498—
LMArena German1493—
LMArena Japanese1502—
LMArena Korean1459—
LMArena Russian1515—
LMArena Spanish1485—

Instruction Following Not comparable

Gemini 3.8 Flash: 78.0 (#13), Phi-3.5-MoE: —

Instruction Following benchmarks
BenchmarkGemini 3.8 FlashPhi-3.5-MoE
LMArena Instruction Following1490—

Long Context Not comparable

Gemini 3.8 Flash: 46.3 (#24), Phi-3.5-MoE: —

Long Context benchmarks
BenchmarkGemini 3.8 FlashPhi-3.5-MoE
LMArena Longer Query1508—

Writing & Preference Not comparable

Gemini 3.8 Flash: 72.2 (#15), Phi-3.5-MoE: —

Writing & Preference benchmarks
BenchmarkGemini 3.8 FlashPhi-3.5-MoE
LMArena Text1499—
LMArena Creative Writing1492—
EQ-Bench Creative Writing1748—
LMArena Multi-Turn1501—

Frequently asked questions

Is Gemini 3.8 Flash better than Phi-3.5-MoE?

Gemini 3.8 Flash has enough public results to be ranked (#11); Phi-3.5-MoE does not yet, so treat this comparison as directional.

How many benchmarks do Gemini 3.8 Flash and Phi-3.5-MoE share?

0 benchmarks have published results for both models. Gemini 3.8 Flash has 50 scored results on Noometry and Phi-3.5-MoE has 3.

Related comparisons

Go deeper