76 lines
1.8 KiB
Markdown
76 lines
1.8 KiB
Markdown
|
---
|
||
|
|
license: other
|
||
|
|
library_name: gguf
|
||
|
|
tags:
|
||
|
|
- gguf
|
||
|
|
- llama
|
||
|
|
- llama-3.2
|
||
|
|
- text-generation
|
||
|
|
- local-llm
|
||
|
|
- llama-cpp
|
||
|
|
- lm-studio
|
||
|
|
- ollama
|
||
|
|
- 1b
|
||
|
|
pipeline_tag: text-generation
|
||
|
|
language:
|
||
|
|
- en
|
||
|
|
base_model_relation: quantized
|
||
|
|
base_model:
|
||
|
|
- gss1147/Llama-3.2-OctoThinker-iNano-1B
|
||
|
|
---
|
||
|
|
|
||
|
|
# Llama-3.2-OctoThinker-iNano-1B-GGUF
|
||
|
|
|
||
|
|
## Model Summary
|
||
|
|
|
||
|
|
**Llama-3.2-OctoThinker-iNano-1B-GGUF** is the GGUF quantized release of the main model:
|
||
|
|
|
||
|
|
**Main model repo:**
|
||
|
|
https://huggingface.co/gss1147/Llama-3.2-OctoThinker-iNano-1B
|
||
|
|
|
||
|
|
This repository packages the model for efficient **local inference** in GGUF-compatible runtimes such as **llama.cpp**, **LM Studio**, and similar local tools.
|
||
|
|
|
||
|
|
## GGUF to Main Model Link
|
||
|
|
|
||
|
|
This GGUF repository corresponds to the main model repo:
|
||
|
|
|
||
|
|
**[`gss1147/Llama-3.2-OctoThinker-iNano-1B`](https://huggingface.co/gss1147/Llama-3.2-OctoThinker-iNano-1B)**
|
||
|
|
|
||
|
|
If you want the original non-GGUF model, training/merge details, tokenizer files, and main repository metadata, use the repo above.
|
||
|
|
|
||
|
|
## Available Files
|
||
|
|
|
||
|
|
This GGUF repository currently includes:
|
||
|
|
|
||
|
|
- **Q4_K_M** — 955 MB
|
||
|
|
- **Q5_K_M** — 1.09 GB
|
||
|
|
- **F16** — 3 GB
|
||
|
|
|
||
|
|
## Architecture
|
||
|
|
|
||
|
|
- **Architecture:** llama
|
||
|
|
- **Model size:** 1B params
|
||
|
|
|
||
|
|
## Intended Use
|
||
|
|
|
||
|
|
This model is intended for:
|
||
|
|
|
||
|
|
- local text generation
|
||
|
|
- assistant-style prompting
|
||
|
|
- lightweight reasoning tasks
|
||
|
|
- summarization
|
||
|
|
- simple coding help
|
||
|
|
- offline/local inference workflows
|
||
|
|
|
||
|
|
## Quantization Notes
|
||
|
|
|
||
|
|
Choose the file that best matches your hardware:
|
||
|
|
|
||
|
|
- **Q4_K_M** for smaller size and lighter RAM usage
|
||
|
|
- **Q5_K_M** for a stronger quality-to-size balance
|
||
|
|
- **F16** for the highest-fidelity file in this repo, with much higher memory requirements
|
||
|
|
|
||
|
|
## Example llama.cpp Usage
|
||
|
|
|
||
|
|
```bash
|
||
|
|
llama-cli -m /path/to/Llama-3.2-OctoThinker-iNano-1B.Q4_K_M.gguf -p "Explain recursion in Python with a simple example."
|