Model comparison
MiniMax-M2.5 vs Nemotron 3 Super
Nemotron 3 Super is the stronger model overall, scoring 40.1 to 38.3 on the Noometry Index.
Last verified . 17 shared benchmarks.
MiniMax
NVIDIASummary
- They share 17 benchmarks with published results for both. MiniMax-M2.5 scores higher in 3 categories and Nemotron 3 Super in 5 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in math, where Nemotron 3 Super leads 39.6 to 26.9.
- Nemotron 3 Super is cheaper at $0.08 / $0.45 per million input/output tokens, against $0.30 / $1.20 for MiniMax-M2.5.
- Nemotron 3 Super accepts more context: 262K tokens versus 205K.
Side by side
| MiniMax-M2.5 | Nemotron 3 Super | |
|---|---|---|
| Provider | MiniMax | NVIDIA |
| Noometry Index | 38.3 | 40.1 |
| Released | 2026-02-12 | 2026-03-11 |
| Weights | Open | Open |
| Context window | 205K | 262K |
| Max output | 131K | 131K |
| Input $ / M tokens | $0.30 | $0.08 |
| Output $ / M tokens | $1.20 | $0.45 |
| Results tracked | 33 | 21 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding MiniMax-M2.5 leads
MiniMax-M2.5: 48.1 (#58), Nemotron 3 Super: 39.4 (#158)
| Benchmark | MiniMax-M2.5 | Nemotron 3 Super |
|---|---|---|
| LMArena Coding | 1381 | 1403 |
| ALE-Bench | 618.17 | 213.9 |
| SWE-bench Verified (bash only) | 75.8% | — |
| LMArena WebDev | 1387 | — |
| SWE-bench Multilingual | 68.3% | — |
| SciCode | — | 36% |
| WeirdML | — | 38% |
Agentic & Tool Use Not comparable
MiniMax-M2.5: 30.4 (#77), Nemotron 3 Super: —
| Benchmark | MiniMax-M2.5 | Nemotron 3 Super |
|---|---|---|
| Terminal-Bench | 42.7% | — |
| Vending-Bench 2 | -23.16 | — |
Reasoning Nemotron 3 Super leads
MiniMax-M2.5: 17.5 (#292), Nemotron 3 Super: 18.7 (#276)
| Benchmark | MiniMax-M2.5 | Nemotron 3 Super |
|---|---|---|
| NYT Connections (extended) | 16.8% | 15.4% |
| LMArena Hard Prompts | 1372 | 1386 |
| ARC-AGI-2 | 4.9% | — |
| Kagi LLM Benchmark | 55.2% | — |
| ARC-AGI-1 | 63.7% | — |
| CritPt | — | 3.1% |
| Epoch Capabilities Index | 146.68 | — |
Math Nemotron 3 Super leads
MiniMax-M2.5: 26.9 (#253), Nemotron 3 Super: 39.6 (#101)
| Benchmark | MiniMax-M2.5 | Nemotron 3 Super |
|---|---|---|
| LMArena Math | 1378 | 1380 |
| MathArena Final-Answer Competitions | — | 60.4% |
| ProofBench | 4% | — |
Knowledge Too close to call
MiniMax-M2.5: 39.2 (#135), Nemotron 3 Super: 38.9 (#139)
| Benchmark | MiniMax-M2.5 | Nemotron 3 Super |
|---|---|---|
| LMArena Expert | 1379 | 1398 |
| Vectara Hallucination Rate | 9.1% | — |
Multilingual Nemotron 3 Super leads
MiniMax-M2.5: 47.1 (#152), Nemotron 3 Super: 48.4 (#144)
| Benchmark | MiniMax-M2.5 | Nemotron 3 Super |
|---|---|---|
| LMArena Non-English | 1338 | 1355 |
| LMArena Chinese | 1393 | 1432 |
| LMArena French | 1362 | 1405 |
| LMArena German | 1362 | 1341 |
| LMArena Russian | 1358 | 1336 |
| LMArena Spanish | 1354 | 1417 |
| LMArena Japanese | 1171 | — |
| LMArena Korean | 1232 | — |
Instruction Following Too close to call
MiniMax-M2.5: 71.5 (#148), Nemotron 3 Super: 71.2 (#154)
| Benchmark | MiniMax-M2.5 | Nemotron 3 Super |
|---|---|---|
| LMArena Instruction Following | 1353 | 1347 |
Long Context Nemotron 3 Super leads
MiniMax-M2.5: 37.5 (#216), Nemotron 3 Super: 41.5 (#139)
| Benchmark | MiniMax-M2.5 | Nemotron 3 Super |
|---|---|---|
| LMArena Longer Query | 1366 | 1362 |
| CL-bench | 11.4% | — |
| CL-bench Life | 6.3% | — |
Writing & Preference Nemotron 3 Super leads
MiniMax-M2.5: 53.9 (#153), Nemotron 3 Super: 55.9 (#140)
| Benchmark | MiniMax-M2.5 | Nemotron 3 Super |
|---|---|---|
| LMArena Text | 1359 | 1378 |
| LMArena Creative Writing | 1331 | 1317 |
| LMArena Multi-Turn | 1364 | 1369 |
| EQ-Bench Creative Writing | 1361 | — |
Frequently asked questions
Is MiniMax-M2.5 better than Nemotron 3 Super?
Nemotron 3 Super is the stronger model overall, scoring 40.1 to 38.3 on the Noometry Index.
Which is cheaper, MiniMax-M2.5 or Nemotron 3 Super?
Nemotron 3 Super is cheaper. It lists at $0.08 per million input tokens and $0.45 per million output tokens; MiniMax-M2.5 lists at $0.30 and $1.20.
Is MiniMax-M2.5 or Nemotron 3 Super better for coding?
MiniMax-M2.5 scores higher on coding benchmarks: 48.1 versus 39.4 in the Noometry coding category.
Which has the bigger context window?
Nemotron 3 Super does, with 262K tokens against 205K.
How many benchmarks do MiniMax-M2.5 and Nemotron 3 Super share?
17 benchmarks have published results for both models. MiniMax-M2.5 has 33 scored results on Noometry and Nemotron 3 Super has 21.