| Model profile | ||
|---|---|---|
| Model | Size | Approximate size |
| Nemotron 3 UltraModel source: provider1,048,576 tokens max native context | 550Bactive 55B | 330 GB4-bit quantization |
| Compatible hardware Estimated speeds, not benchmark results: calculated from memory bandwidth and model size. Real results can differ significantly because there is no precise formula for deriving LLM generation speed from hardware specifications alone. | ||||
|---|---|---|---|---|
| Hardware | Memory | Bandwidth | vLLM | Estimated generation |
| DGX H200 | 1128 GBGPU HBM3e | 4800 GB/s | yes | 55 tok/s4-bit quantization |
| DGX Station | 748 GBCoherent Memory | 7100 GB/s | yes | 79 tok/s4-bit quantization |
| ET900N G3 | 748 GBCoherent Memory | 7100 GB/s | yes | 79 tok/s4-bit quantization |
| DGX B200 | 1440 GBGPU HBM3e | 8000 GB/s | yes | 88 tok/s4-bit quantization |
| GB200 NVL72 | 13400 GBGPU HBM3e | 8000 GB/s | yes | 88 tok/s4-bit quantization |
| GB300 NVL72 | 20000 GBGPU HBM3e | 8000 GB/s | yes | 88 tok/s4-bit quantization |
Evaluates Nemotron 3 ULTRA through browser OS, scene generation, OpenCode, flight and 3D printer simulations, C++, Python, niche knowledge, roleplay, and website generation tests.
Evaluates NVIDIA Nemotron 3 Ultra using Inferencer App on M3 Ultra 512GB hardware against state-of-the-art models under the new Open License.
Demonstrates using the nvidia/nemotron-3-ultra-550b-a55b:free model via OpenRouter with agents like Claude Code, OpenCode, Pi, KiloCode, VS Code, T3.code, Hermes, and OpenClaw.
Tests NVIDIA-Nemotron-3-Ultra-550B-A55B with hermes-agent.