Hermes 3

What is Hermes 3?

Hermes 3 is Nous Research's open-weights language model. Released in 2024-08, it ships in 4 sizes (3B, 8B, 70B, 405B) and is licensed Apache 2.0. The most popular use case is uncensored chat with strong function calling.

VRAM and hardware

The smallest 3B size needs at least 2 GB VRAM; the largest 405B needs around 2 GB.

Size	Min VRAM (Q4_K_M)	Recommended VRAM	Tokens/sec on 3090
3B	2 GB	3 GB	~31 tok/s
8B	6 GB	9 GB	~24 tok/s
70B	42 GB	63 GB	~14 tok/s
405B	240 GB	360 GB	~10 tok/s

How to run Hermes 3 locally

Option 1: Ollama (simplest)

ollama pull hermes-3
ollama run hermes-3

Option 2: Mullama (production)

mullama pull hermes-3
mullama run hermes-3

Option 3: llama.cpp (CLI)

# Download a GGUF from Hugging Face (search "hermes-3 gguf")
./llama-cli -m hermes-3.Q4_K_M.gguf -p "Hello, AI!"

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

from mullama import Model, Context
model = Model.load("hermes-3.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 Hermes 3

• function calling
• uncensored chat
• agentic

Hardware it fits on

• Apple Silicon (8GB): 3B in MLX
• Apple Silicon (16GB): 8B in MLX
• Apple Silicon (32GB): 70B in MLX or GGUF
• Apple Silicon (64GB+): 405B in MLX
• RTX 3090 (24GB): 70B at Q4_K_M
• RTX 4090 (24GB): same as 3090, ~30% faster
• RTX 5090 (32GB): 70B at Q8_0
• 2× RTX 6000 Ada (96GB total): 405B at Q4_K_M

Related models in the Hermes family

This is the only Hermes-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