Model comparison
DeepSeek-R1 vs MiniMax-M2
DeepSeek-R1 is the stronger model overall, scoring 42.3 to 37.4 on the Noometry Index. MiniMax-M2 costs 1.7× less per token, which makes it the better buy when DeepSeek-R1's lead doesn't matter for your workload.
Last verified . 16 shared benchmarks.
DeepSeek
MiniMaxSummary
- They share 16 benchmarks with published results for both. DeepSeek-R1 scores higher in 8 categories and MiniMax-M2 in 1 category; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where DeepSeek-R1 leads 61.4 to 53.0.
- The biggest single-benchmark swing is Kagi LLM Benchmark: 69.4% for DeepSeek-R1 and 57.8% for MiniMax-M2.
- MiniMax-M2 is cheaper at $0.30 / $1.20 per million input/output tokens, against $0.50 / $2.15 for DeepSeek-R1.
- MiniMax-M2 accepts more context: 205K tokens versus 164K.
- MiniMax-M2 has downloadable open weights; the other is API-only.
Side by side
| DeepSeek-R1 | MiniMax-M2 | |
|---|---|---|
| Provider | DeepSeek | MiniMax |
| Noometry Index | 42.3 | 37.4 |
| Released | 2025-01-20 | 2025-10-27 |
| Weights | Proprietary | Open |
| Context window | 164K | 205K |
| Max output | 64K | 131K |
| Input $ / M tokens | $0.50 | $0.30 |
| Output $ / M tokens | $2.15 | $1.20 |
| Results tracked | 52 | 21 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek-R1 leads
DeepSeek-R1: 46.3 (#68), MiniMax-M2: 39.3 (#159)
| Benchmark | DeepSeek-R1 | MiniMax-M2 |
|---|---|---|
| LMArena Coding | 1427 | 1370 |
| SWE-bench Verified (bash only) | — | 61% |
| Aider Polyglot | 71.4% | — |
| LMArena WebDev | — | 1297 |
| SciCode | 35.7% | — |
| WeirdML | 41.6% | — |
| LiveBench Coding | 66.7% | — |
| ALE-Bench | 804.12 | — |
| AlgoTune | 1.7 | — |
Agentic & Tool Use DeepSeek-R1 leads
DeepSeek-R1: 30.7 (#75), MiniMax-M2: 25.1 (#109)
| Benchmark | DeepSeek-R1 | MiniMax-M2 |
|---|---|---|
| Terminal-Bench | — | 30% |
| DeepResearch Bench | 35.1% | — |
| BALROG | 34.9% | — |
| METR Time Horizons | 53.8% | — |
| Vending-Bench 2 | — | 160.6 |
Reasoning Too close to call
DeepSeek-R1: 18.6 (#278), MiniMax-M2: 19.4 (#258)
| Benchmark | DeepSeek-R1 | MiniMax-M2 |
|---|---|---|
| Kagi LLM Benchmark | 69.4% | 57.8% |
| LMArena Hard Prompts | 1416 | 1357 |
| ARC-AGI-2 | 1.3% | — |
| SimpleBench | 40.8% | — |
| NYT Connections (extended) | — | 14.8% |
| ARC-AGI-1 | 21.2% | — |
| CritPt | 1.1% | — |
| LiveBench Reasoning | 83.2% | — |
| LiveBench Data Analysis | 69.8% | — |
| Epoch Capabilities Index | 141.29 | — |
| ForecastBench | 60 | — |
| LiveBench | 71.6% | — |
Math DeepSeek-R1 leads
DeepSeek-R1: 43.8 (#79), MiniMax-M2: 37.3 (#160)
| Benchmark | DeepSeek-R1 | MiniMax-M2 |
|---|---|---|
| LMArena Math | 1400 | 1352 |
| OTIS Mock AIME 2024-2025 | 66.4% | — |
| Omni-MATH | 42.4% | — |
| LiveBench Math | 80.7% | — |
| MATH Level 5 | 96.6% | — |
Knowledge DeepSeek-R1 leads
DeepSeek-R1: 44.5 (#87), MiniMax-M2: 37.0 (#163)
| Benchmark | DeepSeek-R1 | MiniMax-M2 |
|---|---|---|
| LMArena Expert | 1394 | 1337 |
| GPQA Diamond | 76.3% | — |
| MMLU-Pro | 79.3% | — |
| Confabulations | 12.7% | — |
| Vectara Hallucination Rate | 11.3% | — |
| GPQA (HELM) | 66.6% | — |
Multilingual DeepSeek-R1 leads
DeepSeek-R1: 52.4 (#85), MiniMax-M2: 45.3 (#171)
| Benchmark | DeepSeek-R1 | MiniMax-M2 |
|---|---|---|
| LMArena Non-English | 1412 | 1313 |
| LMArena Chinese | 1442 | 1366 |
| LMArena French | 1417 | 1335 |
| LMArena German | 1404 | 1355 |
| LMArena Russian | 1423 | 1331 |
| LMArena Spanish | 1411 | 1326 |
| LMArena Japanese | 1391 | — |
| LMArena Korean | 1360 | — |
Instruction Following DeepSeek-R1 leads
DeepSeek-R1: 72.0 (#143), MiniMax-M2: 70.2 (#166)
| Benchmark | DeepSeek-R1 | MiniMax-M2 |
|---|---|---|
| LMArena Instruction Following | 1382 | 1328 |
| LiveBench Instruction Following | 80.5% | — |
| IFEval | 78.4% | — |
Long Context DeepSeek-R1 leads
DeepSeek-R1: 45.4 (#36), MiniMax-M2: 40.5 (#153)
| Benchmark | DeepSeek-R1 | MiniMax-M2 |
|---|---|---|
| LMArena Longer Query | 1391 | 1331 |
| Fiction.LiveBench | 75% | — |
Writing & Preference DeepSeek-R1 leads
DeepSeek-R1: 61.4 (#88), MiniMax-M2: 53.0 (#162)
| Benchmark | DeepSeek-R1 | MiniMax-M2 |
|---|---|---|
| LMArena Text | 1428 | 1340 |
| LMArena Creative Writing | 1405 | 1286 |
| LMArena Multi-Turn | 1405 | 1361 |
| Short-Story Creative Writing | 83% | — |
| EQ-Bench Creative Writing | 1500 | — |
| WildBench | 82.8% | — |
| LiveBench Language | 48.5% | — |
Frequently asked questions
Is DeepSeek-R1 better than MiniMax-M2?
DeepSeek-R1 is the stronger model overall, scoring 42.3 to 37.4 on the Noometry Index. MiniMax-M2 costs 1.7× less per token, which makes it the better buy when DeepSeek-R1's lead doesn't matter for your workload.
Which is cheaper, DeepSeek-R1 or MiniMax-M2?
MiniMax-M2 is cheaper. It lists at $0.30 per million input tokens and $1.20 per million output tokens; DeepSeek-R1 lists at $0.50 and $2.15.
Is DeepSeek-R1 or MiniMax-M2 better for coding?
DeepSeek-R1 scores higher on coding benchmarks: 46.3 versus 39.3 in the Noometry coding category.
Which has the bigger context window?
MiniMax-M2 does, with 205K tokens against 164K.
How many benchmarks do DeepSeek-R1 and MiniMax-M2 share?
16 benchmarks have published results for both models. DeepSeek-R1 has 52 scored results on Noometry and MiniMax-M2 has 21.