# DeepSeek-V3.1 vs MiniMax-M3

> MiniMax-M3 is the stronger model overall, scoring 43.5 to 42.2 on the Noometry Index.

- Canonical page: https://noometry.com/compare/deepseek-v3-1-vs-minimax-m3
- Last updated: 2026-10-10
- Shared benchmarks: 22

## Summary

- They share 22 benchmarks with published results for both. DeepSeek-V3.1 scores higher in 0 categories and MiniMax-M3 in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where MiniMax-M3 leads 58.4 to 43.0.
- The biggest single-benchmark swing is LMCA: 24.3% for DeepSeek-V3.1 and 33.7% for MiniMax-M3.
- DeepSeek-V3.1 is cheaper at $0.25 / $0.95 per million input/output tokens, against $0.30 / $1.20 for MiniMax-M3.
- MiniMax-M3 accepts more context: 1M tokens versus 164K.

## Snapshot

| | DeepSeek-V3.1 | MiniMax-M3 |
|---|---|---|
| Provider | DeepSeek | MiniMax |
| Noometry Index | 42.2 | 43.5 |
| Rank | 106 | 81 |
| Context | 164K | 1M |
| Input $/M | $0.25 | $0.30 |
| Output $/M | $0.95 | $1.20 |
| Weights | Open | Open |

## Coding

- DeepSeek-V3.1: 39.8 (#140)
- MiniMax-M3: 41.5 (#111)

| Benchmark | DeepSeek-V3.1 | MiniMax-M3 |
|---|---|---|
| LMArena Coding | 1417 | 1470 |
| FrontierCode | — | 14.7% |
| LMArena WebDev | — | 1482 |
| SciCode | — | 47.1% |
| WeirdML | 38.4% | — |
| ALE-Bench | — | 640.02 |

## Agentic & Tool Use

- DeepSeek-V3.1: —
- MiniMax-M3: 22.5 (#121)

| Benchmark | DeepSeek-V3.1 | MiniMax-M3 |
|---|---|---|
| APEX-Agents | — | 37.7% |
| OSWorld 2.0 | — | 4.6% |
| GBAEval | — | 0.9% |
| Vending-Bench 2 | — | 2,158 |

## Reasoning

- DeepSeek-V3.1: 27.3 (#109)
- MiniMax-M3: 29.3 (#83)

| Benchmark | DeepSeek-V3.1 | MiniMax-M3 |
|---|---|---|
| SimpleBench | 40% | 45.8% |
| LMArena Hard Prompts | 1417 | 1447 |
| DTBench | 82.7% | 78.9% |
| LMCA | 24.3% | 33.7% |
| Epoch Capabilities Index | 139.92 | 147 |
| ForecastBench | 58 | 61.4 |
| Kagi LLM Benchmark | 53.2% | — |
| NYT Connections (extended) | — | 65.1% |
| CritPt | — | 3.7% |
| Chess Puzzles | — | 14% |
| Mystery Game Puzzles | — | 8% |
| Surface Evolver Bench | — | 55% |

## Math

- DeepSeek-V3.1: 36.1 (#125)
- MiniMax-M3: 39.1 (#83)

| Benchmark | DeepSeek-V3.1 | MiniMax-M3 |
|---|---|---|
| LMArena Math | 1420 | 1432 |
| OTIS Mock AIME 2024-2025 | — | 71.1% |
| ProofBench | — | 18% |

## Knowledge

- DeepSeek-V3.1: 43.0 (#86)
- MiniMax-M3: 58.4 (#32)

| Benchmark | DeepSeek-V3.1 | MiniMax-M3 |
|---|---|---|
| LMArena Expert | 1403 | 1462 |
| GPQA Diamond | — | 90.9% |
| Vectara Hallucination Rate | 5.5% | — |

## Multimodal

- DeepSeek-V3.1: —
- MiniMax-M3: 39.5 (#45)

| Benchmark | DeepSeek-V3.1 | MiniMax-M3 |
|---|---|---|
| LMArena Vision | — | 1254 |
| LMArena Document | — | 1435 |

## Multilingual

- DeepSeek-V3.1: 51.8 (#100)
- MiniMax-M3: 53.2 (#70)

| Benchmark | DeepSeek-V3.1 | MiniMax-M3 |
|---|---|---|
| LMArena Non-English | 1400 | 1420 |
| LMArena Chinese | 1470 | 1465 |
| LMArena French | 1447 | 1445 |
| LMArena German | 1412 | 1424 |
| LMArena Japanese | 1374 | 1381 |
| LMArena Korean | 1337 | 1372 |
| LMArena Russian | 1405 | 1430 |
| LMArena Spanish | 1429 | 1431 |

## Instruction Following

- DeepSeek-V3.1: 74.1 (#105)
- MiniMax-M3: 75.7 (#56)

| Benchmark | DeepSeek-V3.1 | MiniMax-M3 |
|---|---|---|
| LMArena Instruction Following | 1400 | 1433 |

## Long Context

- DeepSeek-V3.1: 36.3 (#227)
- MiniMax-M3: 44.2 (#68)

| Benchmark | DeepSeek-V3.1 | MiniMax-M3 |
|---|---|---|
| LMArena Longer Query | 1422 | 1445 |
| Fiction.LiveBench | 52.8% | — |

## Writing & Preference

- DeepSeek-V3.1: 60.5 (#93)
- MiniMax-M3: 62.3 (#79)

| Benchmark | DeepSeek-V3.1 | MiniMax-M3 |
|---|---|---|
| LMArena Text | 1420 | 1433 |
| LMArena Creative Writing | 1401 | 1405 |
| LMArena Multi-Turn | 1408 | 1441 |
| EQ-Bench Creative Writing | 1436 | — |
| EQ-Bench 4 | — | 1150 |

## FAQ

### Is DeepSeek-V3.1 better than MiniMax-M3?

MiniMax-M3 is the stronger model overall, scoring 43.5 to 42.2 on the Noometry Index.

### Which is cheaper, DeepSeek-V3.1 or MiniMax-M3?

DeepSeek-V3.1 is cheaper. It lists at $0.25 per million input tokens and $0.95 per million output tokens; MiniMax-M3 lists at $0.30 and $1.20.

### Is DeepSeek-V3.1 or MiniMax-M3 better for coding?

MiniMax-M3 scores higher on coding benchmarks: 41.5 versus 39.8 in the Noometry coding category.

### Which has the bigger context window?

MiniMax-M3 does, with 1M tokens against 164K.

### How many benchmarks do DeepSeek-V3.1 and MiniMax-M3 share?

22 benchmarks have published results for both models. DeepSeek-V3.1 has 27 scored results on Noometry and MiniMax-M3 has 41.