Model comparison
Deepseek Coder v2 vs MiniMax-M2.1
MiniMax-M2.1 is the stronger model overall, scoring 38.4 to 35.3 on the Noometry Index.
Last verified . 17 shared benchmarks.
DeepSeek
MiniMaxSummary
- They share 17 benchmarks with published results for both. Deepseek Coder v2 scores higher in 1 category and MiniMax-M2.1 in 7 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where MiniMax-M2.1 leads 58.5 to 38.4.
Side by side
| Deepseek Coder v2 | MiniMax-M2.1 | |
|---|---|---|
| Provider | DeepSeek | MiniMax |
| Noometry Index | 35.3 | 38.4 |
| Released | 2024-06-17 | 2025-12-23 |
| Weights | Open | Open |
| Context window | — | 205K |
| Max output | — | 131K |
| Input $ / M tokens | — | $0.30 |
| Output $ / M tokens | — | $1.20 |
| Results tracked | 24 | 22 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding MiniMax-M2.1 leads
Deepseek Coder v2: 37.5 (#178), MiniMax-M2.1: 40.1 (#132)
| Benchmark | Deepseek Coder v2 | MiniMax-M2.1 |
|---|---|---|
| LMArena Coding | 1251 | 1421 |
| LMArena WebDev | — | 1384 |
| BigCodeBench Instruct | 48.2% | — |
| BigCodeBench Complete | 59.7% | — |
| ALE-Bench | — | 623.83 |
| HumanEval+ | 82.3% | — |
| MBPP+ | 75.1% | — |
Agentic & Tool Use Not comparable
Deepseek Coder v2: —, MiniMax-M2.1: 27.7 (#91)
| Benchmark | Deepseek Coder v2 | MiniMax-M2.1 |
|---|---|---|
| Terminal-Bench | — | 36.6% |
Reasoning Deepseek Coder v2 leads
Deepseek Coder v2: 22.6 (#175), MiniMax-M2.1: 15.8 (#301)
| Benchmark | Deepseek Coder v2 | MiniMax-M2.1 |
|---|---|---|
| LMArena Hard Prompts | 1207 | 1410 |
| NYT Connections (extended) | — | 11.2% |
| WinoGrande | 83.7% | — |
Math MiniMax-M2.1 leads
Deepseek Coder v2: 32.3 (#188), MiniMax-M2.1: 35.6 (#142)
| Benchmark | Deepseek Coder v2 | MiniMax-M2.1 |
|---|---|---|
| LMArena Math | 1241 | 1396 |
| GSM8K | 94.5% | — |
Knowledge MiniMax-M2.1 leads
Deepseek Coder v2: 31.7 (#209), MiniMax-M2.1: 37.7 (#143)
| Benchmark | Deepseek Coder v2 | MiniMax-M2.1 |
|---|---|---|
| LMArena Expert | 1181 | 1432 |
| Vectara Hallucination Rate | — | 11.8% |
| ARC (AI2) Challenge | 64.3% | — |
Multilingual MiniMax-M2.1 leads
Deepseek Coder v2: 36.4 (#235), MiniMax-M2.1: 50.2 (#123)
| Benchmark | Deepseek Coder v2 | MiniMax-M2.1 |
|---|---|---|
| LMArena Non-English | 1182 | 1378 |
| LMArena Chinese | 1201 | 1430 |
| LMArena French | 1184 | 1405 |
| LMArena German | 1164 | 1380 |
| LMArena Japanese | 1126 | 1287 |
| LMArena Korean | 1104 | 1297 |
| LMArena Russian | 1188 | 1386 |
| LMArena Spanish | 1153 | 1396 |
Instruction Following MiniMax-M2.1 leads
Deepseek Coder v2: 61.9 (#237), MiniMax-M2.1: 74.0 (#108)
| Benchmark | Deepseek Coder v2 | MiniMax-M2.1 |
|---|---|---|
| LMArena Instruction Following | 1180 | 1399 |
Long Context MiniMax-M2.1 leads
Deepseek Coder v2: 37.0 (#219), MiniMax-M2.1: 43.2 (#98)
| Benchmark | Deepseek Coder v2 | MiniMax-M2.1 |
|---|---|---|
| LMArena Longer Query | 1219 | 1416 |
Writing & Preference MiniMax-M2.1 leads
Deepseek Coder v2: 38.4 (#249), MiniMax-M2.1: 58.5 (#114)
| Benchmark | Deepseek Coder v2 | MiniMax-M2.1 |
|---|---|---|
| LMArena Text | 1191 | 1392 |
| LMArena Creative Writing | 1120 | 1362 |
| LMArena Multi-Turn | 1176 | 1396 |
Frequently asked questions
Is Deepseek Coder v2 better than MiniMax-M2.1?
MiniMax-M2.1 is the stronger model overall, scoring 38.4 to 35.3 on the Noometry Index.
Is Deepseek Coder v2 or MiniMax-M2.1 better for coding?
MiniMax-M2.1 scores higher on coding benchmarks: 40.1 versus 37.5 in the Noometry coding category.
How many benchmarks do Deepseek Coder v2 and MiniMax-M2.1 share?
17 benchmarks have published results for both models. Deepseek Coder v2 has 24 scored results on Noometry and MiniMax-M2.1 has 22.