Nanite-Labs/ smoky

3 Downloads Updated 6 days ago

ollama run Nanite-Labs/smoky:q4_k_m

curl http://localhost:11434/api/chat \
  -d '{
    "model": "Nanite-Labs/smoky:q4_k_m",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

from ollama import chat

response = chat(
    model='Nanite-Labs/smoky:q4_k_m',
    messages=[{'role': 'user', 'content': 'Hello!'}],
)
print(response.message.content)

import ollama from 'ollama'

const response = await ollama.chat({
  model: 'Nanite-Labs/smoky:q4_k_m',
  messages: [{role: 'user', content: 'Hello!'}],
})
console.log(response.message.content)

Models

Name

1 model

Size / Usage

Context

Input

smoky:q4_k_m

808MB · 128K context window · Text · 6 days ago

smoky:q4_k_m

808MB

128K

Text

Readme

smoky

Q4_K_M quant of the merged nanites-smoky-1b-chat model — a Llama-3.2-1B-Instruct fine-tune that plays an old resident of the Great Smoky Mountains of the 1930s. You talk to it; the old-timer answers in period voice with regional vocabulary.

ollama run Nanite-Labs/smoky:q4_k_m "How was life growing up around here?"

What it does

Answers modern-English questions about Smokies life (work, family, food, music, religion, hardships, the mountains) in 1–3 sentences in a 1930s regional voice (“warn’t”, “pore”, “y’all”, “holler”, “crick”, “taters”, “a-going”). Small 1B fine-tune on ~93 curated Q&A — a convincing character sketch, not a history source.

What it doesn’t do

Not an authoritative history source. Training Q&A was distilled from Joseph Sargent Hall’s 1939 field recordings; “facts” reflect speakers’ lived experience, not academic history.
No audio. For speech, pair with the CosyVoice2 voices Nanite-Labs/nanites-smoky-speech-earl / -ethel on HuggingFace.

Training

Base: Llama-3.2-1B-Instruct · Method: QLoRA r=16 · Data: 84 train / 9 val Q&A, hand-authored seed + teacher-distilled (local ollama teacher) from the Hall transcripts (used locally only, not redistributed).

Provenance & license

Adapter + recipe: Apache 2.0 (Nanites Laboratories). Base model: Meta Llama 3.2 Community License.
Full weights + LoRA adapter: https://huggingface.co/Nanite-Labs/nanites-smoky-1b-chat · GGUF: https://huggingface.co/Nanite-Labs/nanites-smoky-1b-chat-gguf

# smoky

Q4_K_M quant of the merged **nanites-smoky-1b-chat** model — a Llama-3.2-1B-Instruct fine-tune that plays an old resident of the Great Smoky Mountains of the 1930s. You talk to it; the old-timer answers in period voice with regional vocabulary.

```bash
ollama run Nanite-Labs/smoky:q4_k_m "How was life growing up around here?"
```

## What it does

Answers modern-English questions about Smokies life (work, family, food, music, religion, hardships, the mountains) in 1–3 sentences in a 1930s regional voice ("warn't", "pore", "y'all", "holler", "crick", "taters", "a-going"). Small 1B fine-tune on ~93 curated Q&A — a convincing character sketch, not a history source.

## What it doesn't do

- Not an authoritative history source. Training Q&A was distilled from Joseph Sargent Hall's 1939 field recordings; "facts" reflect speakers' lived experience, not academic history.
- No audio. For speech, pair with the CosyVoice2 voices `Nanite-Labs/nanites-smoky-speech-earl` / `-ethel` on HuggingFace.

## Training

- Base: Llama-3.2-1B-Instruct · Method: QLoRA r=16 · Data: 84 train / 9 val Q&A, hand-authored seed + teacher-distilled (local ollama teacher) from the Hall transcripts (used locally only, not redistributed).

## Provenance & license

- Adapter + recipe: Apache 2.0 (Nanites Laboratories). Base model: Meta Llama 3.2 Community License.
- Full weights + LoRA adapter: https://huggingface.co/Nanite-Labs/nanites-smoky-1b-chat · GGUF: https://huggingface.co/Nanite-Labs/nanites-smoky-1b-chat-gguf

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