Compare / head-to-head
NVIDIA Nemotron 3 Super 120B-A12BvsPhi-4 Reasoning
NVIDIA Nemotron 3 Super 120B-A12B leads 2 of 2 shared benchmarks. NVIDIA Nemotron 3 Super 120B-A12B has the larger context window (1M tokens).
Benchmarks from cited public sources; pricing from official pages; status from official provider feeds.
Shared benchmarks
2 – 0
NVIDIA Nemotron 3 Super 120B-A12B leads
Cheaper per token
—
list price, input + output
Larger context
NVIDIA Nemotron 3 Super 120B-A12B
1M tokens
Provider uptime (30d)
100% · —
NVIDIA · Microsoft
NVIDIA Nemotron 3 Super 120B-A12B
NVIDIA · released 2026-03-11
text
- Context
- 1M
- Max out
- —
- Input /1M
- —
- Output /1M
- —
- Cached /1M
- —
- Scores
- 10 · 8 core
Phi-4 Reasoning
Microsoft · released 2025-04-30
text
- Context
- 33K
- Max out
- 33K
- Input /1M
- —
- Output /1M
- —
- Cached /1M
- —
- Scores
- 4 · 4 core
Quality
Benchmark matrix
2 shared · 8 only NVIDIA Nemotron 3 Super 120B-A12B · 2 only Phi-4 Reasoning| Benchmark | NVIDIA Nemotron 3 Super 120B-A12B | Phi-4 Reasoning | Δ | Edge |
|---|---|---|---|---|
| Reported by both · 2 | ||||
| MMLU-Pro | 83.73% ↗ | 74.3% ↗ | +9.4 pt | NVIDIA Nemotron 3 Super 120B-A12B |
| AIME 2025 | 90.21% ↗ | 62.9% ↗ | +27.3 pt | NVIDIA Nemotron 3 Super 120B-A12B |
| Only NVIDIA Nemotron 3 Super 120B-A12B reports · 8 | ||||
| SWE-bench Verified | 60.47% ↗ | not reported | — | — |
| Humanity's Last Exam (no tools) | 18.26% ↗ | not reported | — | — |
| LMArena Elo | 1377.9 ↗ | not reported | — | — |
| GPQA (no tools) | 79.23% ↗ | not reported | — | — |
| GPQA (with tools) | 82.7% ↗ | not reported | — | — |
| Humanity's Last Exam (with tools) | 22.82% ↗ | not reported | — | — |
| LiveCodeBench | 81.19% ↗ | not reported | — | — |
| Terminal-Bench Core 2.0 | 31% ↗ | not reported | — | — |
| Only Phi-4 Reasoning reports · 2 | ||||
| GPQA Diamond | not reported | 65.8% ↗ | — | — |
| HumanEval+ | not reported | 92.9% ↗ | — | — |
Scores tagged "epoch" or "matharena" are independent runs, used only where the lab has not published its own; ⚠ marks rows MathArena flags as released after the competition. Higher is better on every row. Δ is NVIDIA Nemotron 3 Super 120B-A12B minus Phi-4 Reasoning in the benchmark's own unit. "Not reported" means the lab has not published that figure; it is not a zero. ↗ opens the source.
Specs & pricing
Side by side
Official pricing pages and model cards| Spec | NVIDIA Nemotron 3 Super 120B-A12B | Phi-4 Reasoning | Edge |
|---|---|---|---|
| Context window | 1M tokens | 33K tokens | NVIDIA Nemotron 3 Super 120B-A12B |
| Max output | — | 33K tokens | — |
| Input price / 1M | — | — | — |
| Output price / 1M | — | — | — |
| Cached input / 1M | — | — | — |
| Input + output / 1M Lower is cheaper. List prices; batch, tool and regional fees excluded. | — | — | — |
| Modalities | text | text | Tie |
| Released | 2026-03-11 | 2025-04-30 | — |
| Cited benchmark scores | 10 | 4 | — |
Reliability
Provider status
Last 60 days · refreshed daily on this page · live on /statusMore matchups
NVIDIA Nemotron 3 Super 120B-A12B vs …
Models sharing the most benchmarksMore matchups
Phi-4 Reasoning vs …
Models sharing the most benchmarksBuilt by Respan
Which one wins on your data?
Public benchmarks are a starting point. Run NVIDIA Nemotron 3 Super 120B-A12B and Phi-4 Reasoning on your own prompts with Respan evals, or route to either through one gateway key with automatic failover.