Nemotron 3 33B
By NVIDIA · United States
Updated 2026-07-13
Overview
NVIDIA's dense 33B model targeting balanced chat, code, and reasoning workloads. Fits a single RTX 4090 at Q4 with a 128k context window.
When to pick this model
- Single-GPU local deployment on a 24GB card (RTX 4090/3090) at Q4
- Mixed workloads spanning chat, code generation, and step-by-step reasoning
- Long-document analysis up to 128k tokens
- Self-hosted alternative to mid-tier API models when data must stay on-prem
VRAM requirements by quantization
| Quantization | VRAM required |
|---|---|
| Q4_K_M (recommended) | 19 GB |
| Q5_K_M | 23 GB |
| Q8_0 | 35 GB |
| FP16 (no quantization) | 66 GB |
VRAM figures include model weights plus a typical 8k KV cache and ~600 MB runtime overhead (Ollama / llama.cpp baseline). Add headroom for higher context lengths.
In practice, Nemotron 3 33B wants a 24 GB card at Q4_K_M (19 GB). Stepping up to Q8_0 nearly doubles the footprint to 35 GB, and unquantized FP16 weights take 66 GB — plan your GPU around the Q4 or Q5 figure unless you specifically need the higher fidelity.
Without a GPU, Nemotron 3 33B needs roughly 43 GB of system RAM to run on CPU via llama.cpp or Ollama — workable for background jobs, but far slower than GPU inference. Throughput estimates from our compatibility engine: around 3 tokens/sec on entry-level GPUs, on the order of 12 tokens/sec on a mid-range card, and up to 30 tokens/sec on high-end hardware — assuming the chosen quantization fully fits in VRAM.
What hardware do you need
The table below matches Nemotron 3 33B to common GPU memory tiers, using the highest-fidelity quantization that fully fits each card class. Spilling layers to system RAM works but costs most of the speed, so size your card to the quantization you actually want to run.
| GPU memory | Example cards | Best fit for Nemotron 3 33B |
|---|---|---|
| 8 GB | RTX 5070 Laptop, RTX 5060, RTX 5060 Ti 8GB | Does not fit — needs 19 GB at Q4_K_M |
| 12 GB | RTX 5070, RTX 5070 Ti Laptop, RTX 4080 Laptop | Does not fit — needs 19 GB at Q4_K_M |
| 16 GB | RTX 5080, RTX 4080 Super, Radeon RX 9070 XT | Does not fit — needs 19 GB at Q4_K_M |
| 24 GB | RTX 4090, Radeon RX 7900 XTX, RTX 5090 Laptop | Q5_K_M (23 GB used) |
| 32 GB | RTX 5090 | Q5_K_M (23 GB used) |
Which GPU should you buy to run Nemotron 3 33B?
To run Nemotron 3 33B locally at Q4, you need ~19 GB of VRAM. The best value for this is a RTX 4090 (24 GB VRAM).
As an Amazon Associate, BestLLMfor earns from qualifying purchases, at no extra cost to you. It does not influence our independent rankings.
Strengths
- Dense 33B sized to saturate a 24GB consumer GPU at Q4
- 128k context handles long codebases and reports
- RLHF tuned for reasoning and code, not just chat
- Open weights backed by NVIDIA's research stack
Limitations
- NVIDIA Open Model License has commercial terms worth reviewing carefully
- Gated on Hugging Face (click-through access required)
- Dense 33B is heavier than comparable MoE alternatives at inference
Typical workloads
In our catalog grid, Nemotron 3 33B is filed under Reasoning, Code Generation, Production Agents — the use cases where its size/quality trade-off makes the most sense. Its tags translate to concrete workloads: code generation and review (pair it with an editor integration like Continue.dev or Cline); multi-step reasoning and math-flavoured tasks.
The 125k-token context window covers long chats and mid-sized documents, though very large retrieval workloads will need chunking. It ships under the NVIDIA Open Model License license — commercial use is generally possible but read the specific terms before embedding it in a product.
Architecture & training
Architecture: Dense Transformer · 33B parameters · 128k context
Training: NVIDIA Nemotron family, RLHF alignment focused on reasoning and code.
A solid single-GPU workhorse for teams that want strong reasoning and code on a 4090 without depending on an API.
Quick start
ollama run nemotron3Or use the open-source MCP server to query this model from Claude Desktop, Cursor, or any MCP-compatible client.
Similar models worth comparing
Frequently asked questions
How much VRAM does Nemotron 3 33B need?
At the recommended Q4_K_M quantization, Nemotron 3 33B needs about 19 GB of VRAM. Q8_0 takes 35 GB, and unquantized FP16 weights take 66 GB.
Can Nemotron 3 33B run without a GPU?
Yes — with roughly 43 GB of system RAM it runs CPU-only through llama.cpp or Ollama. Expect a fraction of GPU speed, which is fine for background or batch jobs but slow for interactive chat.
What context window does Nemotron 3 33B support?
Nemotron 3 33B supports a 125k-token context window (128,000 tokens).
Can I use Nemotron 3 33B commercially?
Nemotron 3 33B ships under the NVIDIA Open Model License license. Commercial use is generally permitted subject to its terms — review the license text before shipping a product.
How fast is Nemotron 3 33B on consumer hardware?
Our compatibility engine estimates on the order of 12 tokens/sec on a mid-range GPU and up to 30 tokens/sec on high-end cards, assuming the quantization fully fits in VRAM.
Which quantization of Nemotron 3 33B should I download first?
Start with Q4_K_M (19 GB) — the standard size/quality sweet spot. Step up to Q5_K_M or Q8_0 only if you have VRAM headroom. On a 24 GB card you can run up to Q5_K_M.