# DeepSeek-V3 vs MiniMax-M2.1

> DeepSeek-V3 and MiniMax-M2.1 score almost the same on the Noometry Index (39.2 vs 38.4), so choose on price, context window or the category you care about most.

- Canonical page: https://noometry.com/compare/deepseek-v3-vs-minimax-m2-1
- Last updated: 2026-10-10
- Shared benchmarks: 18

## Summary

- They share 18 benchmarks with published results for both. DeepSeek-V3 scores higher in 2 categories and MiniMax-M2.1 in 6 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in long context, where MiniMax-M2.1 leads 43.2 to 34.0.
- The biggest single-benchmark swing is Vectara Hallucination Rate: 6.1% for DeepSeek-V3 and 11.8% for MiniMax-M2.1.
- DeepSeek-V3 is cheaper at $0.24 / $0.90 per million input/output tokens, against $0.30 / $1.20 for MiniMax-M2.1.
- MiniMax-M2.1 accepts more context: 205K tokens versus 164K.

## Snapshot

| | DeepSeek-V3 | MiniMax-M2.1 |
|---|---|---|
| Provider | DeepSeek | MiniMax |
| Noometry Index | 39.2 | 38.4 |
| Rank | 157 | 177 |
| Context | 164K | 205K |
| Input $/M | $0.24 | $0.30 |
| Output $/M | $0.90 | $1.20 |
| Weights | Open | Open |

## Coding

- DeepSeek-V3: 41.9 (#99)
- MiniMax-M2.1: 40.1 (#132)

| Benchmark | DeepSeek-V3 | MiniMax-M2.1 |
|---|---|---|
| LMArena Coding | 1368 | 1421 |
| Aider Polyglot | 55.1% | — |
| LMArena WebDev | — | 1384 |
| SciCode | 35.8% | — |
| WeirdML | 36.1% | — |
| BigCodeBench Instruct | 50% | — |
| LiveBench Coding | 70.9% | — |
| BigCodeBench Complete | 62.2% | — |
| ALE-Bench | — | 623.83 |
| HumanEval+ | 86.6% | — |
| MBPP+ | 73% | — |

## Agentic & Tool Use

- DeepSeek-V3: —
- MiniMax-M2.1: 27.7 (#91)

| Benchmark | DeepSeek-V3 | MiniMax-M2.1 |
|---|---|---|
| Terminal-Bench | — | 36.6% |
| METR Time Horizons | 49.6% | — |

## Reasoning

- DeepSeek-V3: 20.0 (#226)
- MiniMax-M2.1: 15.8 (#301)

| Benchmark | DeepSeek-V3 | MiniMax-M2.1 |
|---|---|---|
| LMArena Hard Prompts | 1365 | 1410 |
| SimpleBench | 27.2% | — |
| Kagi LLM Benchmark | 52.3% | — |
| NYT Connections (extended) | — | 11.2% |
| CritPt | 0% | — |
| LiveBench Reasoning | 65.8% | — |
| DTBench | 64.8% | — |
| LiveBench Data Analysis | 60.9% | — |
| LMCA | 15.5% | — |
| BIG-Bench Hard | 87.5% | — |
| Epoch Capabilities Index | 135.95 | — |
| ForecastBench | 59.1 | — |
| HellaSwag | 88.9% | — |
| LiveBench | 66.9% | — |
| PIQA | 84.7% | — |
| WinoGrande | 85.2% | — |

## Math

- DeepSeek-V3: 30.4 (#208)
- MiniMax-M2.1: 35.6 (#142)

| Benchmark | DeepSeek-V3 | MiniMax-M2.1 |
|---|---|---|
| LMArena Math | 1373 | 1396 |
| OTIS Mock AIME 2024-2025 | 37.8% | — |
| Omni-MATH | 40.3% | — |
| LiveBench Math | 73.5% | — |
| MATH Level 5 | 75.5% | — |
| FrontierMath (Feb 2025 set) | 1.7% | — |

## Knowledge

- DeepSeek-V3: 37.5 (#147)
- MiniMax-M2.1: 37.7 (#143)

| Benchmark | DeepSeek-V3 | MiniMax-M2.1 |
|---|---|---|
| Vectara Hallucination Rate | 6.1% | 11.8% |
| LMArena Expert | 1352 | 1432 |
| GPQA Diamond | 67.6% | — |
| MMLU-Pro | 72.3% | — |
| Confabulations | 26.1% | — |
| GPQA (HELM) | 53.8% | — |
| ARC (AI2) Challenge | 95.3% | — |
| MMLU | 87.2% | — |
| TriviaQA | 82.9% | — |

## Multilingual

- DeepSeek-V3: 48.8 (#138)
- MiniMax-M2.1: 50.2 (#123)

| Benchmark | DeepSeek-V3 | MiniMax-M2.1 |
|---|---|---|
| LMArena Non-English | 1358 | 1378 |
| LMArena Chinese | 1390 | 1430 |
| LMArena French | 1386 | 1405 |
| LMArena German | 1374 | 1380 |
| LMArena Japanese | 1333 | 1287 |
| LMArena Korean | 1321 | 1297 |
| LMArena Russian | 1372 | 1386 |
| LMArena Spanish | 1358 | 1396 |

## Instruction Following

- DeepSeek-V3: 73.0 (#125)
- MiniMax-M2.1: 74.0 (#108)

| Benchmark | DeepSeek-V3 | MiniMax-M2.1 |
|---|---|---|
| LMArena Instruction Following | 1345 | 1399 |
| LiveBench Instruction Following | 81.5% | — |
| IFEval | 83.2% | — |

## Long Context

- DeepSeek-V3: 34.0 (#248)
- MiniMax-M2.1: 43.2 (#98)

| Benchmark | DeepSeek-V3 | MiniMax-M2.1 |
|---|---|---|
| LMArena Longer Query | 1353 | 1416 |
| Fiction.LiveBench | 50% | — |

## Writing & Preference

- DeepSeek-V3: 57.6 (#124)
- MiniMax-M2.1: 58.5 (#114)

| Benchmark | DeepSeek-V3 | MiniMax-M2.1 |
|---|---|---|
| LMArena Text | 1375 | 1392 |
| LMArena Creative Writing | 1364 | 1362 |
| LMArena Multi-Turn | 1389 | 1396 |
| Short-Story Creative Writing | 77% | — |
| EQ-Bench Creative Writing | 1472 | — |
| WildBench | 83% | — |
| LiveBench Language | 49.1% | — |

## FAQ

### Is DeepSeek-V3 better than MiniMax-M2.1?

DeepSeek-V3 and MiniMax-M2.1 score almost the same on the Noometry Index (39.2 vs 38.4), so choose on price, context window or the category you care about most.

### Which is cheaper, DeepSeek-V3 or MiniMax-M2.1?

DeepSeek-V3 is cheaper. It lists at $0.24 per million input tokens and $0.90 per million output tokens; MiniMax-M2.1 lists at $0.30 and $1.20.

### Is DeepSeek-V3 or MiniMax-M2.1 better for coding?

DeepSeek-V3 scores higher on coding benchmarks: 41.9 versus 40.1 in the Noometry coding category.

### Which has the bigger context window?

MiniMax-M2.1 does, with 205K tokens against 164K.

### How many benchmarks do DeepSeek-V3 and MiniMax-M2.1 share?

18 benchmarks have published results for both models. DeepSeek-V3 has 60 scored results on Noometry and MiniMax-M2.1 has 22.