fauxpaslife/theprint-10B-MoE-A3B-0126-Q6

fauxpaslife/ theprint-10B-MoE-A3B-0126-Q6_K:latest

11 Downloads Updated 1 month ago

Llama Mix of Specialists with a small footprint.

tools

ollama run fauxpaslife/theprint-10B-MoE-A3B-0126-Q6_K

curl http://localhost:11434/api/chat \
  -d '{
    "model": "fauxpaslife/theprint-10B-MoE-A3B-0126-Q6_K",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

from ollama import chat

response = chat(
    model='fauxpaslife/theprint-10B-MoE-A3B-0126-Q6_K',
    messages=[{'role': 'user', 'content': 'Hello!'}],
)
print(response.message.content)

import ollama from 'ollama'

const response = await ollama.chat({
  model: 'fauxpaslife/theprint-10B-MoE-A3B-0126-Q6_K',
  messages: [{role: 'user', content: 'Hello!'}],
})
console.log(response.message.content)

Details

Updated 1 month ago

1 month ago

5885b86aa1f8 · 7.8GB

model

archllama

parameters9.55B

quantizationQ6_K

7.8GB

params

{ "stop": [ "<|start_header_id|>", "<|end_header_id|>", "<|eot_id|>"

65B

template

{{- if or .System .Tools }}<|start_header_id|>system<|end_header_id|> {{- if .System }} {{ .System }

1.5kB

Readme

New Report bugs or version requests.

### theprint-10B-MoE-A3B Model: https://huggingface.co/theprint/theprint-10B-MoE-A3B-0126
GGUF: https://huggingface.co/mradermacher/theprint-10B-MoE-A3B-0126-GGUF

A Mixture of Experts model built on Llama 3.2 3B, combining four specialized fine-tunes with a general-purpose model. Architecture

Base model: theprint/GeneralChat-Llama3.2-3B
Gate mode: Hidden
Dtype: bfloat16
Experts: 4

Experts

The model uses a hidden gate mechanism to route inputs to the most relevant expert(s) based on the content of the prompt. Each expert was fine-tuned for its domain before being merged into this MoE architecture using mergekit. Usage

Compatible with any Llama 3.2 inference setup. No special configuration required — the routing happens automatically.

**New** 
[Report bugs or version requests](https://deltakitsune.org/report-a-bug/).

**### theprint-10B-MoE-A3B**
![Screenshot 2026-09-26 111623.png](/assets/fauxpaslife/theprint-10B-MoE-A3B-0126-Q6_K/eff60404-b0e4-4b4f-84ee-35bab02e59ec)
**Model**: https://huggingface.co/theprint/theprint-10B-MoE-A3B-0126 <br>
**GGUF**: https://huggingface.co/mradermacher/theprint-10B-MoE-A3B-0126-GGUF

A Mixture of Experts model built on Llama 3.2 3B, combining four specialized fine-tunes with a general-purpose model.
Architecture

Base model: theprint/GeneralChat-Llama3.2-3B
    Gate mode: Hidden
    Dtype: bfloat16
    Experts: 4

Experts
![Screenshot 2026-09-26 111836.png](/assets/fauxpaslife/theprint-10B-MoE-A3B-0126-Q6_K/f106b9a3-d3a5-48ce-9cdd-95f77c5ccbd9)

Compatible with any Llama 3.2 inference setup. No special configuration required — the routing happens automatically.

![Screenshot 2026-09-26 112510.png](/assets/fauxpaslife/theprint-10B-MoE-A3B-0126-Q6_K/27994803-8675-4210-97da-75f9936bcdea)

Paste, drop or click to upload images (.png, .jpeg, .jpg, .svg, .gif)