# DeepSeek LLM 67B vs MiniMax-M2.5

> MiniMax-M2.5 is the stronger model overall, scoring 38.1 to 24.8 on the Noometry Index.

- Canonical page: https://noometry.com/compare/deepseek-llm-67b-vs-minimax-m2-5
- Last updated: 2026-10-10
- Shared benchmarks: 11

## Summary

- They share 11 benchmarks with published results for both. DeepSeek LLM 67B scores higher in 0 categories and MiniMax-M2.5 in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where MiniMax-M2.5 leads 38.6 to 7.1.

## Snapshot

| | DeepSeek LLM 67B | MiniMax-M2.5 |
|---|---|---|
| Provider | DeepSeek | MiniMax |
| Noometry Index | 24.8 | 38.1 |
| Rank | 341 | 181 |
| Context | — | 205K |
| Input $/M | — | $0.30 |
| Output $/M | — | $1.20 |
| Weights | Open | Open |

## Coding

- DeepSeek LLM 67B: 31.3 (#273)
- MiniMax-M2.5: 47.9 (#53)

| Benchmark | DeepSeek LLM 67B | MiniMax-M2.5 |
|---|---|---|
| LMArena Coding | 1096 | 1380 |
| SWE-bench Verified (bash only) | — | 75.8% |
| LMArena WebDev | — | 1386 |
| SWE-bench Multilingual | — | 68.3% |
| ALE-Bench | — | 618.17 |

## Agentic & Tool Use

- DeepSeek LLM 67B: —
- MiniMax-M2.5: 30.2 (#70)

| Benchmark | DeepSeek LLM 67B | MiniMax-M2.5 |
|---|---|---|
| Terminal-Bench | — | 42.7% |
| Vending-Bench 2 | — | -23.16 |

## Reasoning

- DeepSeek LLM 67B: 15.9 (#300)
- MiniMax-M2.5: 17.6 (#281)

| Benchmark | DeepSeek LLM 67B | MiniMax-M2.5 |
|---|---|---|
| LMArena Hard Prompts | 1070 | 1372 |
| Epoch Capabilities Index | 110.54 | 146.68 |
| ARC-AGI-2 | — | 4.9% |
| Kagi LLM Benchmark | — | 55.2% |
| NYT Connections (extended) | — | 16.8% |
| ARC-AGI-1 | — | 63.7% |
| Chess Puzzles | 0% | — |

## Math

- DeepSeek LLM 67B: 8.1 (#318)
- MiniMax-M2.5: 25.8 (#242)

| Benchmark | DeepSeek LLM 67B | MiniMax-M2.5 |
|---|---|---|
| LMArena Math | 1108 | 1378 |
| OTIS Mock AIME 2024-2025 | 0.8% | — |
| ProofBench | — | 4% |
| MATH Level 5 | 6.4% | — |

## Knowledge

- DeepSeek LLM 67B: 7.1 (#308)
- MiniMax-M2.5: 38.6 (#132)

| Benchmark | DeepSeek LLM 67B | MiniMax-M2.5 |
|---|---|---|
| GPQA Diamond | 24.6% | — |
| Vectara Hallucination Rate | — | 9.1% |
| LMArena Expert | — | 1378 |

## Multilingual

- DeepSeek LLM 67B: 29.5 (#262)
- MiniMax-M2.5: 47.3 (#147)

| Benchmark | DeepSeek LLM 67B | MiniMax-M2.5 |
|---|---|---|
| LMArena Non-English | 1073 | 1338 |
| LMArena Chinese | 1132 | 1392 |
| LMArena French | — | 1363 |
| LMArena German | — | 1361 |
| LMArena Japanese | — | 1169 |
| LMArena Korean | — | 1232 |
| LMArena Russian | — | 1359 |
| LMArena Spanish | — | 1354 |

## Instruction Following

- DeepSeek LLM 67B: 55.7 (#272)
- MiniMax-M2.5: 71.7 (#143)

| Benchmark | DeepSeek LLM 67B | MiniMax-M2.5 |
|---|---|---|
| LMArena Instruction Following | 1079 | 1353 |

## Long Context

- DeepSeek LLM 67B: 33.1 (#260)
- MiniMax-M2.5: 37.4 (#211)

| Benchmark | DeepSeek LLM 67B | MiniMax-M2.5 |
|---|---|---|
| LMArena Longer Query | 1092 | 1367 |
| CL-bench | — | 11.4% |
| CL-bench Life | — | 6.3% |

## Writing & Preference

- DeepSeek LLM 67B: 31.8 (#277)
- MiniMax-M2.5: 54.1 (#149)

| Benchmark | DeepSeek LLM 67B | MiniMax-M2.5 |
|---|---|---|
| LMArena Text | 1105 | 1359 |
| LMArena Creative Writing | 1067 | 1331 |
| LMArena Multi-Turn | 1082 | 1364 |
| EQ-Bench Creative Writing | — | 1361 |

## FAQ

### Is DeepSeek LLM 67B better than MiniMax-M2.5?

MiniMax-M2.5 is the stronger model overall, scoring 38.1 to 24.8 on the Noometry Index.

### Is DeepSeek LLM 67B or MiniMax-M2.5 better for coding?

MiniMax-M2.5 scores higher on coding benchmarks: 47.9 versus 31.3 in the Noometry coding category.

### How many benchmarks do DeepSeek LLM 67B and MiniMax-M2.5 share?

11 benchmarks have published results for both models. DeepSeek LLM 67B has 15 scored results on Noometry and MiniMax-M2.5 has 33.