# Nemotron 3 Ultra vs Phi 3 Medium 4k Instruct

> Nemotron 3 Ultra is the stronger model overall, scoring 44.7 to 32.4 on the Noometry Index.

- Canonical page: https://noometry.com/compare/nemotron-3-ultra-vs-phi-3-medium-4k-instruct
- Last updated: 2026-10-10
- Shared benchmarks: 16

## Summary

- They share 16 benchmarks with published results for both. Nemotron 3 Ultra scores higher in 8 categories and Phi 3 Medium 4k Instruct in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Nemotron 3 Ultra leads 66.3 to 34.3.

## Snapshot

| | Nemotron 3 Ultra | Phi 3 Medium 4k Instruct |
|---|---|---|
| Provider | NVIDIA | Microsoft |
| Noometry Index | 44.7 | 32.4 |
| Rank | 71 | 246 |
| Context | 262K | — |
| Input $/M | $0.60 | — |
| Output $/M | $2.40 | — |
| Weights | Open | Open |

## Coding

- Nemotron 3 Ultra: 37.7 (#174)
- Phi 3 Medium 4k Instruct: 32.3 (#263)

| Benchmark | Nemotron 3 Ultra | Phi 3 Medium 4k Instruct |
|---|---|---|
| LMArena Coding | 1468 | 1130 |
| FrontierCode | 13.6% | — |
| SciCode | 40.3% | — |
| WeirdML | 43.5% | — |

## Reasoning

- Nemotron 3 Ultra: 30.3 (#78)
- Phi 3 Medium 4k Instruct: 20.8 (#214)

| Benchmark | Nemotron 3 Ultra | Phi 3 Medium 4k Instruct |
|---|---|---|
| LMArena Hard Prompts | 1452 | 1127 |
| CritPt | 3.1% | — |
| Chess Puzzles | 12% | — |
| Mystery Game Puzzles | 20% | — |
| DTBench | 90.1% | — |
| LMCA | 36.9% | — |
| Epoch Capabilities Index | 146.2 | — |

## Math

- Nemotron 3 Ultra: 34.4 (#160)
- Phi 3 Medium 4k Instruct: 30.9 (#202)

| Benchmark | Nemotron 3 Ultra | Phi 3 Medium 4k Instruct |
|---|---|---|
| LMArena Math | 1456 | 1173 |
| OTIS Mock AIME 2024-2025 | 86.7% | — |
| ProofBench | 2% | — |

## Knowledge

- Nemotron 3 Ultra: 52.5 (#56)
- Phi 3 Medium 4k Instruct: 29.6 (#226)

| Benchmark | Nemotron 3 Ultra | Phi 3 Medium 4k Instruct |
|---|---|---|
| LMArena Expert | 1472 | 1108 |
| GPQA Diamond | 85.4% | — |

## Multilingual

- Nemotron 3 Ultra: 53.7 (#61)
- Phi 3 Medium 4k Instruct: 30.8 (#258)

| Benchmark | Nemotron 3 Ultra | Phi 3 Medium 4k Instruct |
|---|---|---|
| LMArena Non-English | 1427 | 1094 |
| LMArena Chinese | 1498 | 1108 |
| LMArena French | 1475 | 1109 |
| LMArena German | 1471 | 1101 |
| LMArena Korean | 1387 | 954 |
| LMArena Russian | 1415 | 1145 |
| LMArena Spanish | 1452 | 1095 |
| LMArena Japanese | — | 1042 |

## Instruction Following

- Nemotron 3 Ultra: 74.9 (#84)
- Phi 3 Medium 4k Instruct: 57.8 (#263)

| Benchmark | Nemotron 3 Ultra | Phi 3 Medium 4k Instruct |
|---|---|---|
| LMArena Instruction Following | 1417 | 1114 |

## Long Context

- Nemotron 3 Ultra: 43.9 (#77)
- Phi 3 Medium 4k Instruct: 33.9 (#250)

| Benchmark | Nemotron 3 Ultra | Phi 3 Medium 4k Instruct |
|---|---|---|
| LMArena Longer Query | 1437 | 1120 |

## Writing & Preference

- Nemotron 3 Ultra: 66.3 (#35)
- Phi 3 Medium 4k Instruct: 34.3 (#267)

| Benchmark | Nemotron 3 Ultra | Phi 3 Medium 4k Instruct |
|---|---|---|
| LMArena Text | 1444 | 1138 |
| LMArena Creative Writing | 1392 | 1106 |
| LMArena Multi-Turn | 1412 | 1088 |
| EQ-Bench Creative Writing | 1692 | — |

## FAQ

### Is Nemotron 3 Ultra better than Phi 3 Medium 4k Instruct?

Nemotron 3 Ultra is the stronger model overall, scoring 44.7 to 32.4 on the Noometry Index.

### Is Nemotron 3 Ultra or Phi 3 Medium 4k Instruct better for coding?

Nemotron 3 Ultra scores higher on coding benchmarks: 37.7 versus 32.3 in the Noometry coding category.

### How many benchmarks do Nemotron 3 Ultra and Phi 3 Medium 4k Instruct share?

16 benchmarks have published results for both models. Nemotron 3 Ultra has 29 scored results on Noometry and Phi 3 Medium 4k Instruct has 17.