fauxpaslife/mira-12B

fauxpaslife/ mira-12B

9 Downloads Updated 2 weeks ago

A 12B chat model that talks like a person, not like an assistant. Short when you're short, warm when you're down, straight to the point when you're busy. No bullet-point essays, no "Great question!", no em dashes. It thinks before it answers.

tools thinking

ollama run fauxpaslife/mira-12B

curl http://localhost:11434/api/chat \
  -d '{
    "model": "fauxpaslife/mira-12B",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

from ollama import chat

response = chat(
    model='fauxpaslife/mira-12B',
    messages=[{'role': 'user', 'content': 'Hello!'}],
)
print(response.message.content)

import ollama from 'ollama'

const response = await ollama.chat({
  model: 'fauxpaslife/mira-12B',
  messages: [{role: 'user', content: 'Hello!'}],
})
console.log(response.message.content)

Applications

Claude Code ollama launch claude --model fauxpaslife/mira-12B

OpenCode ollama launch opencode --model fauxpaslife/mira-12B

Hermes Agent ollama launch hermes --model fauxpaslife/mira-12B

OpenClaw ollama launch openclaw --model fauxpaslife/mira-12B

Models

View all →

Name

1 model

Size / Usage

Context

Input

mira-12B:latest

7.4GB · 256K context window · Text · 2 weeks ago

mira-12B:latest

7.4GB

256K

Text

Readme

Sample, that its functional.

HF Author: beyoru/mira-12b
HF GGUF: mradermacher/mira-12b-GGUF
Q4_K_M Quant

Mira 12B

A 12B chat model that talks like a person, not like an assistant. Short when you’re short, warm when you’re down, straight to the point when you’re busy. No bullet-point essays, no “Great question!”, no em dashes. It thinks before it answers, and its thinking reads like someone mulling it over, not a to-do list.
It keeps the base model’s brains: math, code and tool calling stay where they were, and it runs on a single consumer GPU or in your device

Important –> The License.

Citation:

@misc{mira12b2026,
  title        = {Mira 12B: a chat model that talks like a person},
  author       = {beyoru},
  year         = {2026},
  howpublished = {\url{https://huggingface.co/beyoru/mira-12b}},
  note         = {Fine-tuned from Gemma 4 12B}
}

Sample, that its functional.

![image.png](/assets/fauxpaslife/mira-12B/426bb19d-e481-4791-a8a8-c0979a7e30c5)

**HF Author**: [beyoru/mira-12b](https://huggingface.co/beyoru/mira-12b)
 **HF GGUF**: [mradermacher/mira-12b-GGUF](https://huggingface.co/mradermacher/mira-12b-GGUF)
 `Q4_K_M` Quant 

![](/assets/fauxpaslife/mira-12B/17af1db4-4bfd-4838-b9ec-6113f0ecf962)

### Mira 12B 
A 12B chat model that talks like a person, not like an assistant. Short when you're short, warm when you're down, straight to the point when you're busy. No bullet-point essays, no "Great question!", no em dashes. It thinks before it answers, and its thinking reads like someone mulling it over, not a to-do list.
 
It keeps the base model's brains: math, code and tool calling stay where they were, and it runs on a single consumer GPU or in your device
 
![image.png](/assets/fauxpaslife/mira-12B/2abb630c-00c0-4a25-a4bd-edf0cb8d857b)

![image.png](/assets/fauxpaslife/mira-12B/790dc4fc-ffe6-4dde-9dbd-b41283707bc4)

Important --> The License. 
![image.png](/assets/fauxpaslife/mira-12B/a5a93f6d-eecc-4519-980b-f396bdfdd06b)

Citation:
```
@misc{mira12b2026,
  title        = {Mira 12B: a chat model that talks like a person},
  author       = {beyoru},
  year         = {2026},
  howpublished = {\url{https://huggingface.co/beyoru/mira-12b}},
  note         = {Fine-tuned from Gemma 4 12B}
}
```

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