Noometry

Model comparison

Gemini 3.1 Pro Preview vs Phi-4 Mini

Gemini 3.1 Pro Preview is the stronger model overall, scoring 56.7 to 30.9 on the Noometry Index. Phi-4 Mini costs 34× less per token, which makes it the better buy when Gemini 3.1 Pro Preview's lead doesn't matter for your workload.

Last verified . 3 shared benchmarks.

Gemini 3.1 Pro Preview Google

56.7

Rank #23 Confirmed

Phi-4 Mini Microsoft

30.9

Rank #283 Reported

Summary

  • They share 3 benchmarks with published results for both. Gemini 3.1 Pro Preview scores higher in 3 categories and Phi-4 Mini in 0 categories; 3 gaps are clear of the uncertainty.
  • The widest gap is in reasoning, where Gemini 3.1 Pro Preview leads 71.7 to 22.4.
  • The biggest single-benchmark swing is SciCode: 58.9% for Gemini 3.1 Pro Preview and 10.8% for Phi-4 Mini.
  • Phi-4 Mini is cheaper at $0.075 / $0.30 per million input/output tokens, against $2 / $12 for Gemini 3.1 Pro Preview.
  • Gemini 3.1 Pro Preview accepts more context: 1.05M tokens versus 128K.
  • Phi-4 Mini has downloadable open weights; the other is API-only.

Side by side

Gemini 3.1 Pro PreviewPhi-4 Mini
ProviderGoogleMicrosoft
Noometry Index56.730.9
Released2026-02-192024-12-11
WeightsProprietaryOpen
Context window1.05M128K
Max output66K4K
Input $ / M tokens$2$0.075
Output $ / M tokens$12$0.30
Results tracked713

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

Category by category

Coding Gemini 3.1 Pro Preview leads

Gemini 3.1 Pro Preview: 42.5 (#99), Phi-4 Mini: 28.1 (#317)

BenchmarkGemini 3.1 Pro PreviewPhi-4 Mini
SciCode58.9%10.8%
SWE-bench Verified75.6%—
DeepSWE11.7%—
LMArena WebDev1447—
GSO22.6%—
WeirdML72.1%—
LMArena Coding1484—
MirrorCode8.9%—
ALE-Bench1,161—
AlgoTune2.02—

Agentic & Tool Use Not comparable

Gemini 3.1 Pro Preview: 37.7 (#34), Phi-4 Mini: —

BenchmarkGemini 3.1 Pro PreviewPhi-4 Mini
Terminal-Bench80.2%—
APEX-Agents35.3%—
τ²-bench Banking26%—
DeepResearch Bench47.8%—
PostTrainBench22%—
BALROG57%—
ExploitBench26.1%—
GBAEval0.8%—
GDP.pdf17%—
LMArena Search1211—
METR Time Horizons77%—
Vending-Bench 23,774—

Reasoning Gemini 3.1 Pro Preview leads

Gemini 3.1 Pro Preview: 71.7 (#12), Phi-4 Mini: 22.4 (#195)

BenchmarkGemini 3.1 Pro PreviewPhi-4 Mini
CritPt17.7%0%
ARC-AGI-277.1%—
SimpleBench79.6%—
NYT Connections (extended)97.4%—
ARC-AGI-198%—
Chess Puzzles55%—
EnigmaEval36.8%—
Thematic Generalization79.4%—
EBR-Bench14.3%—
LMArena Hard Prompts1485—
Mystery Game Puzzles34%—
DTBench97.1%—
LMCA53.8%—
Epoch Capabilities Index154.77—
ForecastBench59—

Math Not comparable

Gemini 3.1 Pro Preview: 62.1 (#34), Phi-4 Mini: —

BenchmarkGemini 3.1 Pro PreviewPhi-4 Mini
FrontierMath (Tiers 1-3)59.6%—
FrontierMath Tier 426.8%—
MathArena Final-Answer Competitions86.5%—
OTIS Mock AIME 2024-202595.6%—
ProofBench26%—
LMArena Math1485—
FrontierMath (Feb 2025 set)36.9%—
FrontierMath Tier 4 (v1)16.7%—

Knowledge Gemini 3.1 Pro Preview leads

Gemini 3.1 Pro Preview: 71.8 (#3), Phi-4 Mini: 25.3 (#262)

BenchmarkGemini 3.1 Pro PreviewPhi-4 Mini
Vectara Hallucination Rate10.4%23.5%
GPQA Diamond94.4%—
Humanity's Last Exam46.4%—
SimpleQA Verified73.5%—
LMArena Expert1485—

Multimodal Not comparable

Gemini 3.1 Pro Preview: 37.9 (#69), Phi-4 Mini: —

BenchmarkGemini 3.1 Pro PreviewPhi-4 Mini
LMArena Vision1296—
Blueprint-Bench 226.5%—
Furniture Assembly26.7%—
LMArena Document1444—

Multilingual Not comparable

Gemini 3.1 Pro Preview: 57.0 (#12), Phi-4 Mini: —

BenchmarkGemini 3.1 Pro PreviewPhi-4 Mini
LMArena Non-English1477—
LMArena Chinese1529—
LMArena French1487—
LMArena German1491—
LMArena Japanese1493—
LMArena Korean1455—
LMArena Russian1498—
LMArena Spanish1479—

Instruction Following Not comparable

Gemini 3.1 Pro Preview: 77.0 (#32), Phi-4 Mini: —

BenchmarkGemini 3.1 Pro PreviewPhi-4 Mini
LMArena Instruction Following1466—

Long Context Not comparable

Gemini 3.1 Pro Preview: 47.4 (#18), Phi-4 Mini: —

BenchmarkGemini 3.1 Pro PreviewPhi-4 Mini
CL-bench20.8%—
CL-bench Life16.9%—
LMArena Longer Query1483—

Writing & Preference Not comparable

Gemini 3.1 Pro Preview: 66.1 (#37), Phi-4 Mini: —

BenchmarkGemini 3.1 Pro PreviewPhi-4 Mini
LMArena Text1481—
LMArena Creative Writing1482—
EQ-Bench Creative Writing1491—
EQ-Bench 41142—
LMArena Multi-Turn1488—

Frequently asked questions

Is Gemini 3.1 Pro Preview better than Phi-4 Mini?

Gemini 3.1 Pro Preview is the stronger model overall, scoring 56.7 to 30.9 on the Noometry Index. Phi-4 Mini costs 34× less per token, which makes it the better buy when Gemini 3.1 Pro Preview's lead doesn't matter for your workload.

Which is cheaper, Gemini 3.1 Pro Preview or Phi-4 Mini?

Phi-4 Mini is cheaper. It lists at $0.075 per million input tokens and $0.30 per million output tokens; Gemini 3.1 Pro Preview lists at $2 and $12.

Is Gemini 3.1 Pro Preview or Phi-4 Mini better for coding?

Gemini 3.1 Pro Preview scores higher on coding benchmarks: 42.5 versus 28.1 in the Noometry coding category.

Which has the bigger context window?

Gemini 3.1 Pro Preview does, with 1.05M tokens against 128K.

How many benchmarks do Gemini 3.1 Pro Preview and Phi-4 Mini share?

3 benchmarks have published results for both models. Gemini 3.1 Pro Preview has 71 scored results on Noometry and Phi-4 Mini has 3.

Related comparisons

Go deeper