Nomic Embed Text

What is Nomic Embed Text?

Nomic Embed Text is Nomic AI's open-weights language model. Released in 2024-02, it ships in 1 size (137M) and is licensed Apache 2.0. The most popular use case is default embedding model for RAG.

VRAM and hardware

The 137M size fits in CPU of VRAM.

Size	Min VRAM (Q4_K_M)	Recommended VRAM	Tokens/sec on 3090
137M	CPU	CPU	~40 tok/s

How to run Nomic Embed Text locally

Option 1: Ollama (simplest)

ollama pull nomic-embed-text
ollama run nomic-embed-text

Option 2: Mullama (production)

mullama pull nomic-embed-text
mullama run nomic-embed-text

Option 3: llama.cpp (CLI)

# Download a GGUF from Hugging Face (search "nomic-embed-text gguf")
./llama-cli -m nomic-embed-text.Q4_K_M.gguf -p "Hello, AI!"

Option 4: Python with Mullama or llama-cpp-python

from mullama import Model, Context
model = Model.load("nomic-embed-text.Q4_K_M.gguf", n_gpu_layers=99)
ctx = Context(model, n_ctx=4096)
print(ctx.generate("Hello, AI!", 256))

What you can build with Nomic Embed Text

• RAG
• retrieval
• semantic search

Hardware it fits on

• Apple Silicon (8GB): 137M in MLX
• Apple Silicon (16GB): 137M in MLX
• Apple Silicon (32GB): 137M in MLX or GGUF
• Apple Silicon (64GB+): 137M in MLX
• RTX 3090 (24GB): 137M at Q4_K_M
• RTX 4090 (24GB): same as 3090, ~30% faster
• RTX 5090 (32GB): 137M at Q8_0
• 2× RTX 6000 Ada (96GB total): 137M at Q4_K_M

Related models in the Nomic family

This is the only Nomic-family model in the catalog right now.

Other models you might consider

• Llama 3.1 (Meta) — 8B, general-purpose chat, agentic workflows, code
• Llama 3.2 (Meta) — 1B, low-end hardware, mobile, edge deployment
• Llama 3.2 Vision (Meta) — 11B, vision + chat on desktop GPUs
• Llama 3.3 (Meta) — 70B, best-in-class 70B for desktops
• Llama 4 (Meta) — Scout 17B (109B MoE), vision + long context (10M tokens)
• Qwen 2.5 (Alibaba) — 0.5B, strongest 7B-32B on consumer hardware