MiniMax-M3 vs Nemotron 3 Ultra 550B A55B: Benchmarks, Pricing & Capabilities Compared
TL;DR — MiniMax-M3 wins for reasoning + cost + long-context · Nemotron 3 Ultra 550B A55B wins for general use.
MiniMax-M3 MiniMax
- Released
- 2026-06-01
- Context window
- 1M tokens
- Input price
- $0.30 / Mtok
- Output price
- $1.20 / Mtok
Nemotron 3 Ultra 550B A55B NVIDIA
- Released
- 2026-06-04
- Context window
- 262K tokens
- Input price
- $0.60 / Mtok
- Output price
- $2.60 / Mtok
Benchmark comparison
| Benchmark | MiniMax-M3 | Nemotron 3 Ultra 550B A55B |
|---|---|---|
| AA Intelligence Index | 54.7 ✓ | 47.7 |
| GPQA Diamond | 92.9% ✓ | 86.7% |
| HLE | 37.1% ✓ | 26.6% |
| IF-Bench | 82.9% ✓ | 81.4% |
| LiveCodeBench Reasoning | 74.0% ✓ | 67.0% |
| SciCode | 45.4% ✓ | 39.9% |
| TAU2-bench | 88.9% ✓ | 83.3% |
| TerminalBench-Hard | 42.4% ✓ | 36.4% |
Pricing comparison
| Metric | MiniMax-M3 | Nemotron 3 Ultra 550B A55B |
|---|---|---|
| Input ($/Mtok) | $0.30 | $0.60 |
| Output ($/Mtok) | $1.20 | $2.60 |
| Cached input ($/Mtok) | — | — |
| Cost per 1M-token roundtrip (1M in + 1M out) | $1.50 | $3.20 |
Context window & modalities
| Attribute | MiniMax-M3 | Nemotron 3 Ultra 550B A55B |
|---|---|---|
| Context window | 1M tokens | 262K tokens |
| Input modalities | text, image, video | text |
| Output modalities | text | text |
| Knowledge cutoff | — | — |
Verdict by use case
Coding
Insufficient data
Basis: SWE-bench
No shared coding benchmark.
Reasoning
→ MiniMax-M3
Basis: GPQA Diamond
MiniMax-M3 92.9% vs Nemotron 3 Ultra 550B A55B 86.7% on GPQA Diamond.
Math
Insufficient data
Basis: MATH / AIME
No shared math benchmark.
Long context
→ MiniMax-M3
Basis: Context window
MiniMax-M3 1M tokens vs Nemotron 3 Ultra 550B A55B 262K tokens.
Cost
→ MiniMax-M3
Basis: Input $/Mtok
MiniMax-M3 $0.3/Mtok vs Nemotron 3 Ultra 550B A55B $0.6/Mtok input.
Changelog & releases
MiniMax-M3
Released 2026-06-01
Nemotron 3 Ultra 550B A55B
Released 2026-06-04