Model comparison
DeepSeek-V3.1-Terminus vs MiniMax-M3
DeepSeek-V3.1-Terminus and MiniMax-M3 score almost the same on the Noometry Index (42.6 vs 43.5), so choose on price, context window or the category you care about most.
Last verified . 15 shared benchmarks.
DeepSeek
MiniMaxSummary
- They share 15 benchmarks with published results for both. DeepSeek-V3.1-Terminus scores higher in 0 categories and MiniMax-M3 in 7 categories; 4 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where MiniMax-M3 leads 29.3 to 25.7.
- The biggest single-benchmark swing is SciCode: 40.6% for DeepSeek-V3.1-Terminus and 47.1% for MiniMax-M3.
- DeepSeek-V3.1-Terminus is cheaper at $0.30 / $1 per million input/output tokens, against $0.30 / $1.20 for MiniMax-M3.
- MiniMax-M3 accepts more context: 1M tokens versus 164K.
Side by side
| DeepSeek-V3.1-Terminus | MiniMax-M3 | |
|---|---|---|
| Provider | DeepSeek | MiniMax |
| Noometry Index | 42.6 | 43.5 |
| Released | 2025-09-22 | 2026-06-01 |
| Weights | Open | Open |
| Context window | 164K | 1M |
| Max output | 66K | 512K |
| Input $ / M tokens | $0.30 | $0.30 |
| Output $ / M tokens | $1 | $1.20 |
| Results tracked | 16 | 41 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
DeepSeek-V3.1-Terminus: 41.5 (#114), MiniMax-M3: 41.5 (#111)
| Benchmark | DeepSeek-V3.1-Terminus | MiniMax-M3 |
|---|---|---|
| SciCode | 40.6% | 47.1% |
| LMArena Coding | 1426 | 1470 |
| ALE-Bench | 745.17 | 640.02 |
| FrontierCode | — | 14.7% |
| LMArena WebDev | — | 1482 |
Agentic & Tool Use Not comparable
DeepSeek-V3.1-Terminus: —, MiniMax-M3: 22.5 (#121)
| Benchmark | DeepSeek-V3.1-Terminus | MiniMax-M3 |
|---|---|---|
| APEX-Agents | — | 37.7% |
| OSWorld 2.0 | — | 4.6% |
| GBAEval | — | 0.9% |
| Vending-Bench 2 | — | 2,158 |
Reasoning MiniMax-M3 leads
DeepSeek-V3.1-Terminus: 25.7 (#126), MiniMax-M3: 29.3 (#83)
| Benchmark | DeepSeek-V3.1-Terminus | MiniMax-M3 |
|---|---|---|
| CritPt | 1.7% | 3.7% |
| LMArena Hard Prompts | 1427 | 1447 |
| DTBench | 81.3% | 78.9% |
| LMCA | 28.6% | 33.7% |
| SimpleBench | — | 45.8% |
| Kagi LLM Benchmark | 57.4% | — |
| NYT Connections (extended) | — | 65.1% |
| Chess Puzzles | — | 14% |
| Mystery Game Puzzles | — | 8% |
| Surface Evolver Bench | — | 55% |
| Epoch Capabilities Index | — | 147 |
| ForecastBench | — | 61.4 |
Math MiniMax-M3 leads
DeepSeek-V3.1-Terminus: 35.7 (#138), MiniMax-M3: 39.1 (#83)
| Benchmark | DeepSeek-V3.1-Terminus | MiniMax-M3 |
|---|---|---|
| LMArena Math | 1402 | 1432 |
| OTIS Mock AIME 2024-2025 | — | 71.1% |
| ProofBench | — | 18% |
Knowledge Not comparable
DeepSeek-V3.1-Terminus: —, MiniMax-M3: 58.4 (#32)
| Benchmark | DeepSeek-V3.1-Terminus | MiniMax-M3 |
|---|---|---|
| GPQA Diamond | — | 90.9% |
| LMArena Expert | — | 1462 |
Multimodal Not comparable
DeepSeek-V3.1-Terminus: —, MiniMax-M3: 39.5 (#45)
| Benchmark | DeepSeek-V3.1-Terminus | MiniMax-M3 |
|---|---|---|
| LMArena Vision | — | 1254 |
| LMArena Document | — | 1435 |
Multilingual Too close to call
DeepSeek-V3.1-Terminus: 52.3 (#87), MiniMax-M3: 53.2 (#70)
| Benchmark | DeepSeek-V3.1-Terminus | MiniMax-M3 |
|---|---|---|
| LMArena Non-English | 1407 | 1420 |
| LMArena Russian | 1434 | 1430 |
| LMArena Chinese | — | 1465 |
| LMArena French | — | 1445 |
| LMArena German | — | 1424 |
| LMArena Japanese | — | 1381 |
| LMArena Korean | — | 1372 |
| LMArena Spanish | — | 1431 |
Instruction Following MiniMax-M3 leads
DeepSeek-V3.1-Terminus: 74.4 (#96), MiniMax-M3: 75.7 (#56)
| Benchmark | DeepSeek-V3.1-Terminus | MiniMax-M3 |
|---|---|---|
| LMArena Instruction Following | 1406 | 1433 |
Long Context Too close to call
DeepSeek-V3.1-Terminus: 43.5 (#88), MiniMax-M3: 44.2 (#68)
| Benchmark | DeepSeek-V3.1-Terminus | MiniMax-M3 |
|---|---|---|
| LMArena Longer Query | 1426 | 1445 |
Writing & Preference MiniMax-M3 leads
DeepSeek-V3.1-Terminus: 61.3 (#87), MiniMax-M3: 62.3 (#79)
| Benchmark | DeepSeek-V3.1-Terminus | MiniMax-M3 |
|---|---|---|
| LMArena Text | 1420 | 1433 |
| LMArena Creative Writing | 1404 | 1405 |
| LMArena Multi-Turn | 1411 | 1441 |
| EQ-Bench 4 | — | 1150 |
Frequently asked questions
Is DeepSeek-V3.1-Terminus better than MiniMax-M3?
DeepSeek-V3.1-Terminus and MiniMax-M3 score almost the same on the Noometry Index (42.6 vs 43.5), so choose on price, context window or the category you care about most.
Which is cheaper, DeepSeek-V3.1-Terminus or MiniMax-M3?
DeepSeek-V3.1-Terminus is cheaper. It lists at $0.30 per million input tokens and $1 per million output tokens; MiniMax-M3 lists at $0.30 and $1.20.
Is DeepSeek-V3.1-Terminus or MiniMax-M3 better for coding?
They score almost the same on coding (41.5 vs 41.5); test both on your own repository before choosing.
Which has the bigger context window?
MiniMax-M3 does, with 1M tokens against 164K.
How many benchmarks do DeepSeek-V3.1-Terminus and MiniMax-M3 share?
15 benchmarks have published results for both models. DeepSeek-V3.1-Terminus has 16 scored results on Noometry and MiniMax-M3 has 41.