Noometry

Model comparison

DeepSeek V4.1 Flash vs Gemini 3 Deep Think

DeepSeek V4.1 Flash has enough public results to be ranked (#38); Gemini 3 Deep Think does not yet, so treat this comparison as directional.

Last verified . 1 shared benchmarks.

DeepSeek V4.1 Flash DeepSeek

52.8

Rank #38 Confirmed

Gemini 3 Deep Think Google

58.3

Unranked Sparse

Summary

  • They share 1 benchmark with published results for both. DeepSeek V4.1 Flash scores higher in 0 categories and Gemini 3 Deep Think in 1 category; one gap is clear of the uncertainty.
  • The widest gap is in reasoning, where Gemini 3 Deep Think leads 70.2 to 50.2.
  • The biggest single-benchmark swing is CritPt: 14.3% for DeepSeek V4.1 Flash and 25.7% for Gemini 3 Deep Think.
  • DeepSeek V4.1 Flash has downloadable open weights; the other is API-only.

Side by side

DeepSeek V4.1 FlashGemini 3 Deep Think
ProviderDeepSeekGoogle
Noometry Index52.858.3
Released2026-09-092026-02-12
WeightsOpenProprietary
Context window1M—
Max output393K—
Input $ / M tokens$0.15—
Output $ / M tokens$0.60—
Results tracked373

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

Category by category

Coding Not comparable

DeepSeek V4.1 Flash: 52.9 (#32), Gemini 3 Deep Think: —

BenchmarkDeepSeek V4.1 FlashGemini 3 Deep Think
LMArena WebDev1619—
SciCode51.9%—
LMArena Coding1506—
ALE-Bench1,092—

Agentic & Tool Use Not comparable

DeepSeek V4.1 Flash: 31.2 (#69), Gemini 3 Deep Think: —

BenchmarkDeepSeek V4.1 FlashGemini 3 Deep Think
APEX-Agents39.5%—
GDP.pdf19.8%—

Reasoning Gemini 3 Deep Think leads

DeepSeek V4.1 Flash: 50.2 (#36), Gemini 3 Deep Think: 70.2 (#14)

BenchmarkDeepSeek V4.1 FlashGemini 3 Deep Think
CritPt14.3%25.7%
ARC-AGI-2—84.6%
NYT Connections (extended)89.6%—
ARC-AGI-1—96%
LMArena Hard Prompts1483—
Mystery Game Puzzles43%—
DTBench89.9%—
LMCA47%—
Surface Evolver Bench46.3%—
Epoch Capabilities Index154.9—

Math Not comparable

DeepSeek V4.1 Flash: 66.7 (#25), Gemini 3 Deep Think: —

BenchmarkDeepSeek V4.1 FlashGemini 3 Deep Think
FrontierMath (Tiers 1-3)67.4%—
FrontierMath Tier 426.8%—
OTIS Mock AIME 2024-202598.3%—
ProofBench54%—
LMArena Math1477—

Knowledge Not comparable

DeepSeek V4.1 Flash: 57.9 (#38), Gemini 3 Deep Think: —

BenchmarkDeepSeek V4.1 FlashGemini 3 Deep Think
GPQA Diamond89.8%—
LMArena Expert1506—

Multimodal Not comparable

DeepSeek V4.1 Flash: 39.1 (#61), Gemini 3 Deep Think: —

BenchmarkDeepSeek V4.1 FlashGemini 3 Deep Think
LMArena Vision1277—
Furniture Assembly34.2%—

Multilingual Not comparable

DeepSeek V4.1 Flash: 55.0 (#35), Gemini 3 Deep Think: —

BenchmarkDeepSeek V4.1 FlashGemini 3 Deep Think
LMArena Non-English1448—
LMArena Chinese1497—
LMArena French1452—
LMArena German1484—
LMArena Japanese1412—
LMArena Korean1452—
LMArena Russian1471—
LMArena Spanish1459—

Instruction Following Not comparable

DeepSeek V4.1 Flash: 77.3 (#26), Gemini 3 Deep Think: —

BenchmarkDeepSeek V4.1 FlashGemini 3 Deep Think
LMArena Instruction Following1474—

Long Context Not comparable

DeepSeek V4.1 Flash: 45.2 (#47), Gemini 3 Deep Think: —

BenchmarkDeepSeek V4.1 FlashGemini 3 Deep Think
LMArena Longer Query1475—

Writing & Preference Not comparable

DeepSeek V4.1 Flash: 65.4 (#48), Gemini 3 Deep Think: —

BenchmarkDeepSeek V4.1 FlashGemini 3 Deep Think
LMArena Text1462—
LMArena Creative Writing1435—
EQ-Bench Creative Writing1540—
LMArena Multi-Turn1457—

Frequently asked questions

Is DeepSeek V4.1 Flash better than Gemini 3 Deep Think?

DeepSeek V4.1 Flash has enough public results to be ranked (#38); Gemini 3 Deep Think does not yet, so treat this comparison as directional.

How many benchmarks do DeepSeek V4.1 Flash and Gemini 3 Deep Think share?

1 benchmark has published results for both models. DeepSeek V4.1 Flash has 37 scored results on Noometry and Gemini 3 Deep Think has 3.

Related comparisons

Go deeper