Model comparison
MiniMax-M2.5 vs Phi 3 Mini 4k Instruct
MiniMax-M2.5 is the stronger model overall, scoring 38.3 to 27.9 on the Noometry Index.
Last verified . 17 shared benchmarks.
MiniMax
MicrosoftSummary
- They share 17 benchmarks with published results for both. MiniMax-M2.5 scores higher in 8 categories and Phi 3 Mini 4k Instruct in 0 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where MiniMax-M2.5 leads 53.9 to 27.6.
Side by side
| MiniMax-M2.5 | Phi 3 Mini 4k Instruct | |
|---|---|---|
| Provider | MiniMax | Microsoft |
| Noometry Index | 38.3 | 27.9 |
| Released | 2026-02-12 | 2024-04-23 |
| Weights | Open | Open |
| Context window | 205K | — |
| Max output | 131K | — |
| Input $ / M tokens | $0.30 | — |
| Output $ / M tokens | $1.20 | — |
| Results tracked | 33 | 35 |
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 Mini 4k Instruct: 26.6 (#323)
| Benchmark | MiniMax-M2.5 | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Coding | 1381 | 1093 |
| SWE-bench Verified (bash only) | 75.8% | — |
| LMArena WebDev | 1387 | — |
| SWE-bench Multilingual | 68.3% | — |
| LiveBench Coding | — | 15.5% |
| ALE-Bench | 618.17 | — |
| HumanEval+ | — | 59.1% |
| MBPP+ | — | 54.2% |
Agentic & Tool Use Not comparable
MiniMax-M2.5: 30.4 (#77), Phi 3 Mini 4k Instruct: —
| Benchmark | MiniMax-M2.5 | Phi 3 Mini 4k Instruct |
|---|---|---|
| Terminal-Bench | 42.7% | — |
| Vending-Bench 2 | -23.16 | — |
Reasoning MiniMax-M2.5 leads
MiniMax-M2.5: 17.5 (#292), Phi 3 Mini 4k Instruct: 14.1 (#328)
| Benchmark | MiniMax-M2.5 | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Hard Prompts | 1372 | 1072 |
| ARC-AGI-2 | 4.9% | — |
| Kagi LLM Benchmark | 55.2% | — |
| NYT Connections (extended) | 16.8% | — |
| ARC-AGI-1 | 63.7% | — |
| Chess Puzzles | — | 0% |
| LiveBench Reasoning | — | 26.8% |
| LiveBench Data Analysis | — | 34.7% |
| Adversarial NLI | — | 52.8% |
| BIG-Bench Hard | — | 71.7% |
| Epoch Capabilities Index | 146.68 | — |
| HellaSwag | — | 76.7% |
| LiveBench | — | 22.4% |
| WinoGrande | — | 70.8% |
Math Too close to call
MiniMax-M2.5: 26.9 (#253), Phi 3 Mini 4k Instruct: 26.6 (#257)
| Benchmark | MiniMax-M2.5 | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Math | 1378 | 1111 |
| ProofBench | 4% | — |
| LiveBench Math | — | 15.7% |
Knowledge MiniMax-M2.5 leads
MiniMax-M2.5: 39.2 (#135), Phi 3 Mini 4k Instruct: 28.5 (#246)
| Benchmark | MiniMax-M2.5 | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Expert | 1379 | 1045 |
| Vectara Hallucination Rate | 9.1% | — |
| ARC (AI2) Challenge | — | 84.9% |
| MMLU | — | 68.8% |
| OpenBookQA | — | 88% |
| TriviaQA | — | 64% |
Multilingual MiniMax-M2.5 leads
MiniMax-M2.5: 47.1 (#152), Phi 3 Mini 4k Instruct: 26.3 (#280)
| Benchmark | MiniMax-M2.5 | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Non-English | 1338 | 1021 |
| LMArena Chinese | 1393 | 1021 |
| LMArena French | 1362 | 1076 |
| LMArena German | 1362 | 1044 |
| LMArena Japanese | 1171 | 935 |
| LMArena Korean | 1232 | 905 |
| LMArena Russian | 1358 | 1022 |
| LMArena Spanish | 1354 | 1085 |
Instruction Following MiniMax-M2.5 leads
MiniMax-M2.5: 71.5 (#148), Phi 3 Mini 4k Instruct: 47.7 (#303)
| Benchmark | MiniMax-M2.5 | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Instruction Following | 1353 | 1053 |
| LiveBench Instruction Following | — | 39.1% |
Long Context MiniMax-M2.5 leads
MiniMax-M2.5: 37.5 (#216), Phi 3 Mini 4k Instruct: 31.7 (#276)
| Benchmark | MiniMax-M2.5 | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Longer Query | 1366 | 1044 |
| CL-bench | 11.4% | — |
| CL-bench Life | 6.3% | — |
Writing & Preference MiniMax-M2.5 leads
MiniMax-M2.5: 53.9 (#153), Phi 3 Mini 4k Instruct: 27.6 (#300)
| Benchmark | MiniMax-M2.5 | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Text | 1359 | 1073 |
| LMArena Creative Writing | 1331 | 1037 |
| LMArena Multi-Turn | 1364 | 1018 |
| EQ-Bench Creative Writing | 1361 | — |
| LiveBench Language | — | 9.2% |
Frequently asked questions
Is MiniMax-M2.5 better than Phi 3 Mini 4k Instruct?
MiniMax-M2.5 is the stronger model overall, scoring 38.3 to 27.9 on the Noometry Index.
Is MiniMax-M2.5 or Phi 3 Mini 4k Instruct better for coding?
MiniMax-M2.5 scores higher on coding benchmarks: 48.1 versus 26.6 in the Noometry coding category.
How many benchmarks do MiniMax-M2.5 and Phi 3 Mini 4k Instruct share?
17 benchmarks have published results for both models. MiniMax-M2.5 has 33 scored results on Noometry and Phi 3 Mini 4k Instruct has 35.