Model comparison
MiniMax-M2.7 vs Nemotron 3 Ultra
Nemotron 3 Ultra is the stronger model overall, scoring 42.5 to 37.7 on the Noometry Index. MiniMax-M2.7 costs 2.0× less per token, which makes it the better buy when Nemotron 3 Ultra's lead doesn't matter for your workload.
Last verified . 21 shared benchmarks.
MiniMax
NVIDIASummary
- They share 21 benchmarks with published results for both. MiniMax-M2.7 scores higher in 2 categories and Nemotron 3 Ultra in 7 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Nemotron 3 Ultra leads 52.5 to 37.7.
- The biggest single-benchmark swing is SciCode: 47% for MiniMax-M2.7 and 40.3% for Nemotron 3 Ultra.
- MiniMax-M2.7 is cheaper at $0.30 / $1.20 per million input/output tokens, against $0.60 / $2.40 for Nemotron 3 Ultra.
- Nemotron 3 Ultra accepts more context: 262K tokens versus 205K.
Side by side
| MiniMax-M2.7 | Nemotron 3 Ultra | |
|---|---|---|
| Provider | MiniMax | NVIDIA |
| Noometry Index | 37.7 | 42.5 |
| Released | 2026-03-18 | 2026-06-04 |
| Weights | Open | Open |
| Context window | 205K | 262K |
| Max output | 131K | 128K |
| Input $ / M tokens | $0.30 | $0.60 |
| Output $ / M tokens | $1.20 | $2.40 |
| Results tracked | 30 | 30 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding MiniMax-M2.7 leads
MiniMax-M2.7: 41.8 (#120), Nemotron 3 Ultra: 38.1 (#182)
| Benchmark | MiniMax-M2.7 | Nemotron 3 Ultra |
|---|---|---|
| SciCode | 47% | 40.3% |
| WeirdML | 37% | 43.5% |
| LMArena Coding | 1454 | 1468 |
| FrontierCode | — | 13.6% |
| LMArena WebDev | 1398 | — |
| ALE-Bench | 599.25 | — |
Agentic & Tool Use MiniMax-M2.7 leads
MiniMax-M2.7: 25.1 (#111), Nemotron 3 Ultra: 23.4 (#126)
| Benchmark | MiniMax-M2.7 | Nemotron 3 Ultra |
|---|---|---|
| Terminal-Bench | 45.1% | — |
| APEX-Agents | — | 22.7% |
| ExploitBench | 13.3% | — |
| GBAEval | 0% | — |
Reasoning Nemotron 3 Ultra leads
MiniMax-M2.7: 19.7 (#253), Nemotron 3 Ultra: 31.0 (#82)
| Benchmark | MiniMax-M2.7 | Nemotron 3 Ultra |
|---|---|---|
| CritPt | 0.6% | 3.1% |
| LMArena Hard Prompts | 1422 | 1452 |
| Epoch Capabilities Index | 145.85 | 146.17 |
| NYT Connections (extended) | 24.7% | — |
| Chess Puzzles | — | 12% |
| Thematic Generalization | 39.3% | — |
| Mystery Game Puzzles | — | 20% |
| DTBench | — | 90.1% |
| LMCA | — | 36.9% |
Math Nemotron 3 Ultra leads
MiniMax-M2.7: 25.9 (#263), Nemotron 3 Ultra: 35.3 (#187)
| Benchmark | MiniMax-M2.7 | Nemotron 3 Ultra |
|---|---|---|
| ProofBench | 3% | 2% |
| LMArena Math | 1420 | 1457 |
| OTIS Mock AIME 2024-2025 | — | 86.7% |
Knowledge Nemotron 3 Ultra leads
MiniMax-M2.7: 37.7 (#152), Nemotron 3 Ultra: 52.5 (#61)
| Benchmark | MiniMax-M2.7 | Nemotron 3 Ultra |
|---|---|---|
| LMArena Expert | 1444 | 1472 |
| GPQA Diamond | — | 85.4% |
| Vectara Hallucination Rate | 12.9% | — |
Multilingual Nemotron 3 Ultra leads
MiniMax-M2.7: 50.3 (#123), Nemotron 3 Ultra: 53.5 (#64)
| Benchmark | MiniMax-M2.7 | Nemotron 3 Ultra |
|---|---|---|
| LMArena Non-English | 1382 | 1427 |
| LMArena Chinese | 1441 | 1497 |
| LMArena French | 1421 | 1472 |
| LMArena German | 1398 | 1471 |
| LMArena Korean | 1313 | 1386 |
| LMArena Russian | 1383 | 1417 |
| LMArena Spanish | 1403 | 1454 |
| LMArena Japanese | 1262 | — |
Instruction Following Too close to call
MiniMax-M2.7: 74.1 (#103), Nemotron 3 Ultra: 74.7 (#88)
| Benchmark | MiniMax-M2.7 | Nemotron 3 Ultra |
|---|---|---|
| LMArena Instruction Following | 1405 | 1418 |
Long Context Too close to call
MiniMax-M2.7: 43.3 (#99), Nemotron 3 Ultra: 43.9 (#81)
| Benchmark | MiniMax-M2.7 | Nemotron 3 Ultra |
|---|---|---|
| LMArena Longer Query | 1419 | 1437 |
Writing & Preference Nemotron 3 Ultra leads
MiniMax-M2.7: 58.9 (#112), Nemotron 3 Ultra: 66.3 (#36)
| Benchmark | MiniMax-M2.7 | Nemotron 3 Ultra |
|---|---|---|
| LMArena Text | 1405 | 1445 |
| LMArena Creative Writing | 1354 | 1398 |
| LMArena Multi-Turn | 1412 | 1411 |
| EQ-Bench Creative Writing | — | 1692 |
Frequently asked questions
Is MiniMax-M2.7 better than Nemotron 3 Ultra?
Nemotron 3 Ultra is the stronger model overall, scoring 42.5 to 37.7 on the Noometry Index. MiniMax-M2.7 costs 2.0× less per token, which makes it the better buy when Nemotron 3 Ultra's lead doesn't matter for your workload.
Which is cheaper, MiniMax-M2.7 or Nemotron 3 Ultra?
MiniMax-M2.7 is cheaper. It lists at $0.30 per million input tokens and $1.20 per million output tokens; Nemotron 3 Ultra lists at $0.60 and $2.40.
Is MiniMax-M2.7 or Nemotron 3 Ultra better for coding?
MiniMax-M2.7 scores higher on coding benchmarks: 41.8 versus 38.1 in the Noometry coding category.
Which has the bigger context window?
Nemotron 3 Ultra does, with 262K tokens against 205K.
How many benchmarks do MiniMax-M2.7 and Nemotron 3 Ultra share?
21 benchmarks have published results for both models. MiniMax-M2.7 has 30 scored results on Noometry and Nemotron 3 Ultra has 30.