# Codellama 70b Instruct vs MiniMax-M2.5

> MiniMax-M2.5 is the stronger model overall, scoring 38.1 to 33.3 on the Noometry Index.

- Canonical page: https://noometry.com/compare/codellama-70b-instruct-vs-minimax-m2-5
- Last updated: 2026-10-10
- Shared benchmarks: 4

## Summary

- They share 4 benchmarks with published results for both. Codellama 70b Instruct scores higher in 1 category and MiniMax-M2.5 in 4 categories; 5 gaps are clear of the uncertainty.
- The widest gap is in multilingual, where MiniMax-M2.5 leads 47.3 to 24.9.

## Snapshot

| | Codellama 70b Instruct | MiniMax-M2.5 |
|---|---|---|
| Provider | Meta | MiniMax |
| Noometry Index | 33.3 | 38.1 |
| Rank | 236 | 181 |
| Context | — | 205K |
| Input $/M | — | $0.30 |
| Output $/M | — | $1.20 |
| Weights | Open | Open |

## Coding

- Codellama 70b Instruct: 37.0 (#189)
- MiniMax-M2.5: 47.9 (#53)

| Benchmark | Codellama 70b Instruct | MiniMax-M2.5 |
|---|---|---|
| SWE-bench Verified (bash only) | — | 75.8% |
| LMArena WebDev | — | 1386 |
| SWE-bench Multilingual | — | 68.3% |
| BigCodeBench Instruct | 40.7% | — |
| LMArena Coding | — | 1380 |
| BigCodeBench Complete | 49.6% | — |
| ALE-Bench | — | 618.17 |
| HumanEval+ | 65.9% | — |

## Agentic & Tool Use

- Codellama 70b Instruct: —
- MiniMax-M2.5: 30.2 (#70)

| Benchmark | Codellama 70b Instruct | MiniMax-M2.5 |
|---|---|---|
| Terminal-Bench | — | 42.7% |
| Vending-Bench 2 | — | -23.16 |

## Reasoning

- Codellama 70b Instruct: 19.2 (#241)
- MiniMax-M2.5: 17.6 (#281)

| Benchmark | Codellama 70b Instruct | MiniMax-M2.5 |
|---|---|---|
| LMArena Hard Prompts | 1052 | 1372 |
| ARC-AGI-2 | — | 4.9% |
| Kagi LLM Benchmark | — | 55.2% |
| NYT Connections (extended) | — | 16.8% |
| ARC-AGI-1 | — | 63.7% |
| Epoch Capabilities Index | — | 146.68 |

## Math

- Codellama 70b Instruct: —
- MiniMax-M2.5: 25.8 (#242)

| Benchmark | Codellama 70b Instruct | MiniMax-M2.5 |
|---|---|---|
| ProofBench | — | 4% |
| LMArena Math | — | 1378 |

## Knowledge

- Codellama 70b Instruct: —
- MiniMax-M2.5: 38.6 (#132)

| Benchmark | Codellama 70b Instruct | MiniMax-M2.5 |
|---|---|---|
| Vectara Hallucination Rate | — | 9.1% |
| LMArena Expert | — | 1378 |

## Multilingual

- Codellama 70b Instruct: 24.9 (#283)
- MiniMax-M2.5: 47.3 (#147)

| Benchmark | Codellama 70b Instruct | MiniMax-M2.5 |
|---|---|---|
| LMArena Non-English | 992 | 1338 |
| LMArena Chinese | — | 1392 |
| LMArena French | — | 1363 |
| LMArena German | — | 1361 |
| LMArena Japanese | — | 1169 |
| LMArena Korean | — | 1232 |
| LMArena Russian | — | 1359 |
| LMArena Spanish | — | 1354 |

## Instruction Following

- Codellama 70b Instruct: 52.1 (#288)
- MiniMax-M2.5: 71.7 (#143)

| Benchmark | Codellama 70b Instruct | MiniMax-M2.5 |
|---|---|---|
| LMArena Instruction Following | 1024 | 1353 |

## Long Context

- Codellama 70b Instruct: —
- MiniMax-M2.5: 37.4 (#211)

| Benchmark | Codellama 70b Instruct | MiniMax-M2.5 |
|---|---|---|
| CL-bench | — | 11.4% |
| CL-bench Life | — | 6.3% |
| LMArena Longer Query | — | 1367 |

## Writing & Preference

- Codellama 70b Instruct: 33.6 (#272)
- MiniMax-M2.5: 54.1 (#149)

| Benchmark | Codellama 70b Instruct | MiniMax-M2.5 |
|---|---|---|
| LMArena Text | 1057 | 1359 |
| LMArena Creative Writing | — | 1331 |
| EQ-Bench Creative Writing | — | 1361 |
| LMArena Multi-Turn | — | 1364 |

## FAQ

### Is Codellama 70b Instruct better than MiniMax-M2.5?

MiniMax-M2.5 is the stronger model overall, scoring 38.1 to 33.3 on the Noometry Index.

### Is Codellama 70b Instruct or MiniMax-M2.5 better for coding?

MiniMax-M2.5 scores higher on coding benchmarks: 47.9 versus 37.0 in the Noometry coding category.

### How many benchmarks do Codellama 70b Instruct and MiniMax-M2.5 share?

4 benchmarks have published results for both models. Codellama 70b Instruct has 7 scored results on Noometry and MiniMax-M2.5 has 33.