Model comparison

MiniMax-M2.5 vs Phi 3 Medium 4k Instruct

MiniMax-M2.5 is the stronger model overall, scoring 38.3 to 33.0 on the Noometry Index.

Last verified . 17 shared benchmarks.

MiniMax-M2.5 MiniMax

38.3

Rank #188 Confirmed

Phi 3 Medium 4k Instruct Microsoft

33.0

Rank #250 Confirmed

Summary

  • They share 17 benchmarks with published results for both. MiniMax-M2.5 scores higher in 6 categories and Phi 3 Medium 4k Instruct in 2 categories; 8 gaps are clear of the uncertainty.
  • The widest gap is in writing & preference, where MiniMax-M2.5 leads 53.9 to 34.1.

Side by side

MiniMax-M2.5 and Phi 3 Medium 4k Instruct specifications
MiniMax-M2.5Phi 3 Medium 4k Instruct
ProviderMiniMaxMicrosoft
Noometry Index38.333.0
Released2026-02-12—
WeightsOpenOpen
Context window205K—
Max output131K—
Input $ / M tokens$0.30—
Output $ / M tokens$1.20—
Results tracked3317

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

Category by category

Coding MiniMax-M2.5 leads

MiniMax-M2.5: 48.1 (#58), Phi 3 Medium 4k Instruct: 32.9 (#268)

Coding benchmarks
BenchmarkMiniMax-M2.5Phi 3 Medium 4k Instruct
LMArena Coding13811130
SWE-bench Verified (bash only)75.8%—
LMArena WebDev1387—
SWE-bench Multilingual68.3%—
ALE-Bench618.17—

Agentic & Tool Use Not comparable

MiniMax-M2.5: 30.4 (#77), Phi 3 Medium 4k Instruct: —

Agentic & Tool Use benchmarks
BenchmarkMiniMax-M2.5Phi 3 Medium 4k Instruct
Terminal-Bench42.7%—
Vending-Bench 2-23.16—

Reasoning Phi 3 Medium 4k Instruct leads

MiniMax-M2.5: 17.5 (#292), Phi 3 Medium 4k Instruct: 21.7 (#214)

Reasoning benchmarks
BenchmarkMiniMax-M2.5Phi 3 Medium 4k Instruct
LMArena Hard Prompts13721127
ARC-AGI-24.9%—
Kagi LLM Benchmark55.2%—
NYT Connections (extended)16.8%—
ARC-AGI-163.7%—
Epoch Capabilities Index146.68—

Math Phi 3 Medium 4k Instruct leads

MiniMax-M2.5: 26.9 (#253), Phi 3 Medium 4k Instruct: 33.4 (#202)

Math benchmarks
BenchmarkMiniMax-M2.5Phi 3 Medium 4k Instruct
LMArena Math13781173
ProofBench4%—

Knowledge MiniMax-M2.5 leads

MiniMax-M2.5: 39.2 (#135), Phi 3 Medium 4k Instruct: 30.2 (#229)

Knowledge benchmarks
BenchmarkMiniMax-M2.5Phi 3 Medium 4k Instruct
LMArena Expert13791108
Vectara Hallucination Rate9.1%—

Multilingual MiniMax-M2.5 leads

MiniMax-M2.5: 47.1 (#152), Phi 3 Medium 4k Instruct: 30.6 (#263)

Multilingual benchmarks
BenchmarkMiniMax-M2.5Phi 3 Medium 4k Instruct
LMArena Non-English13381094
LMArena Chinese13931108
LMArena French13621110
LMArena German13621101
LMArena Japanese11711042
LMArena Korean1232954
LMArena Russian13581145
LMArena Spanish13541095

Instruction Following MiniMax-M2.5 leads

MiniMax-M2.5: 71.5 (#148), Phi 3 Medium 4k Instruct: 57.6 (#268)

Instruction Following benchmarks
BenchmarkMiniMax-M2.5Phi 3 Medium 4k Instruct
LMArena Instruction Following13531113

Long Context MiniMax-M2.5 leads

MiniMax-M2.5: 37.5 (#216), Phi 3 Medium 4k Instruct: 34.0 (#255)

Long Context benchmarks
BenchmarkMiniMax-M2.5Phi 3 Medium 4k Instruct
LMArena Longer Query13661120
CL-bench11.4%—
CL-bench Life6.3%—

Writing & Preference MiniMax-M2.5 leads

MiniMax-M2.5: 53.9 (#153), Phi 3 Medium 4k Instruct: 34.1 (#272)

Writing & Preference benchmarks
BenchmarkMiniMax-M2.5Phi 3 Medium 4k Instruct
LMArena Text13591138
LMArena Creative Writing13311107
LMArena Multi-Turn13641088
EQ-Bench Creative Writing1361—

Frequently asked questions

Is MiniMax-M2.5 better than Phi 3 Medium 4k Instruct?

MiniMax-M2.5 is the stronger model overall, scoring 38.3 to 33.0 on the Noometry Index.

Is MiniMax-M2.5 or Phi 3 Medium 4k Instruct better for coding?

MiniMax-M2.5 scores higher on coding benchmarks: 48.1 versus 32.9 in the Noometry coding category.

How many benchmarks do MiniMax-M2.5 and Phi 3 Medium 4k Instruct share?

17 benchmarks have published results for both models. MiniMax-M2.5 has 33 scored results on Noometry and Phi 3 Medium 4k Instruct has 17.

Related comparisons

Go deeper