# Granite 3.0 2b Instruct vs MiniMax-M2.7

> MiniMax-M2.7 is the stronger model overall, scoring 37.4 to 30.2 on the Noometry Index.

- Canonical page: https://noometry.com/compare/granite-3-0-2b-instruct-vs-minimax-m2-7
- Last updated: 2026-10-10
- Shared benchmarks: 12

## Summary

- They share 12 benchmarks with published results for both. Granite 3.0 2b Instruct scores higher in 2 categories and MiniMax-M2.7 in 6 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where MiniMax-M2.7 leads 59.1 to 29.8.

## Snapshot

| | Granite 3.0 2b Instruct | MiniMax-M2.7 |
|---|---|---|
| Provider | IBM | MiniMax |
| Noometry Index | 30.2 | 37.4 |
| Rank | 284 | 188 |
| Context | — | 205K |
| Input $/M | — | $0.30 |
| Output $/M | — | $1.20 |
| Weights | Open | Open |

## Coding

- Granite 3.0 2b Instruct: 27.8 (#311)
- MiniMax-M2.7: 41.4 (#115)

| Benchmark | Granite 3.0 2b Instruct | MiniMax-M2.7 |
|---|---|---|
| LMArena Coding | 1090 | 1454 |
| LMArena WebDev | — | 1397 |
| SciCode | — | 47% |
| WeirdML | — | 37% |
| BigCodeBench Instruct | 20.5% | — |
| ALE-Bench | — | 599.25 |

## Agentic & Tool Use

- Granite 3.0 2b Instruct: —
- MiniMax-M2.7: 24.9 (#103)

| Benchmark | Granite 3.0 2b Instruct | MiniMax-M2.7 |
|---|---|---|
| Terminal-Bench | — | 45.1% |
| ExploitBench | — | 13.3% |
| GBAEval | — | 0% |

## Reasoning

- Granite 3.0 2b Instruct: 19.6 (#234)
- MiniMax-M2.7: 19.0 (#248)

| Benchmark | Granite 3.0 2b Instruct | MiniMax-M2.7 |
|---|---|---|
| LMArena Hard Prompts | 1073 | 1422 |
| NYT Connections (extended) | — | 24.7% |
| CritPt | — | 0.6% |
| Thematic Generalization | — | 39.3% |
| Epoch Capabilities Index | — | 145.84 |

## Math

- Granite 3.0 2b Instruct: 29.7 (#215)
- MiniMax-M2.7: 24.9 (#254)

| Benchmark | Granite 3.0 2b Instruct | MiniMax-M2.7 |
|---|---|---|
| LMArena Math | 1117 | 1420 |
| ProofBench | — | 3% |

## Knowledge

- Granite 3.0 2b Instruct: 28.4 (#238)
- MiniMax-M2.7: 37.1 (#151)

| Benchmark | Granite 3.0 2b Instruct | MiniMax-M2.7 |
|---|---|---|
| LMArena Expert | 1064 | 1443 |
| Vectara Hallucination Rate | — | 12.9% |

## Multilingual

- Granite 3.0 2b Instruct: 27.1 (#273)
- MiniMax-M2.7: 50.5 (#120)

| Benchmark | Granite 3.0 2b Instruct | MiniMax-M2.7 |
|---|---|---|
| LMArena Non-English | 1033 | 1382 |
| LMArena Chinese | 1070 | 1439 |
| LMArena Russian | 1045 | 1383 |
| LMArena French | — | 1422 |
| LMArena German | — | 1398 |
| LMArena Japanese | — | 1261 |
| LMArena Korean | — | 1313 |
| LMArena Spanish | — | 1403 |

## Instruction Following

- Granite 3.0 2b Instruct: 54.2 (#279)
- MiniMax-M2.7: 74.2 (#101)

| Benchmark | Granite 3.0 2b Instruct | MiniMax-M2.7 |
|---|---|---|
| LMArena Instruction Following | 1056 | 1404 |

## Long Context

- Granite 3.0 2b Instruct: 32.5 (#263)
- MiniMax-M2.7: 43.3 (#94)

| Benchmark | Granite 3.0 2b Instruct | MiniMax-M2.7 |
|---|---|---|
| LMArena Longer Query | 1071 | 1419 |

## Writing & Preference

- Granite 3.0 2b Instruct: 29.8 (#287)
- MiniMax-M2.7: 59.1 (#107)

| Benchmark | Granite 3.0 2b Instruct | MiniMax-M2.7 |
|---|---|---|
| LMArena Text | 1080 | 1404 |
| LMArena Creative Writing | 1046 | 1354 |
| LMArena Multi-Turn | 1053 | 1413 |

## FAQ

### Is Granite 3.0 2b Instruct better than MiniMax-M2.7?

MiniMax-M2.7 is the stronger model overall, scoring 37.4 to 30.2 on the Noometry Index.

### Is Granite 3.0 2b Instruct or MiniMax-M2.7 better for coding?

MiniMax-M2.7 scores higher on coding benchmarks: 41.4 versus 27.8 in the Noometry coding category.

### How many benchmarks do Granite 3.0 2b Instruct and MiniMax-M2.7 share?

12 benchmarks have published results for both models. Granite 3.0 2b Instruct has 13 scored results on Noometry and MiniMax-M2.7 has 30.