52 lines
1.3 KiB
Markdown
52 lines
1.3 KiB
Markdown
|
---
|
||
|
|
base_model: LiquidAI/LFM2-2.6B
|
||
|
|
license: apache-2.0
|
||
|
|
language:
|
||
|
|
- en
|
||
|
|
tags:
|
||
|
|
- gguf
|
||
|
|
- llama.cpp
|
||
|
|
- ollama
|
||
|
|
- lfm2
|
||
|
|
- unsloth
|
||
|
|
- conversational
|
||
|
|
- text-generation
|
||
|
|
---
|
||
|
|
|
||
|
|
# MAJESTIC-FIN-R1 GGUF
|
||
|
|
|
||
|
|
MAJESTIC-FIN-R1 is a fine-tuned `LiquidAI/LFM2-2.6B` model exported to GGUF for Ollama, llama.cpp, and lightweight CPU deployment.
|
||
|
|
|
||
|
|
## Available files
|
||
|
|
|
||
|
|
- `MAJESTIC-FIN-R1-F16.gguf`: highest-fidelity GGUF export.
|
||
|
|
- `MAJESTIC-FIN-R1-Q8_0.gguf`: smaller GGUF export for Ollama and free CPU hosting.
|
||
|
|
- `template`: Ollama chat template for this model family.
|
||
|
|
- `params`: default Ollama runtime parameters.
|
||
|
|
- `Modelfile`: local Ollama import file.
|
||
|
|
|
||
|
|
## Run with Ollama from Hugging Face
|
||
|
|
|
||
|
|
```bash
|
||
|
|
ollama run hf.co/EREN121232/MAJESTIC-FIN-R1-gguf:Q8_0
|
||
|
|
```
|
||
|
|
|
||
|
|
## Run with Ollama locally
|
||
|
|
|
||
|
|
1. Download `MAJESTIC-FIN-R1-Q8_0.gguf` and `Modelfile`.
|
||
|
|
2. Keep them in the same folder.
|
||
|
|
3. Run:
|
||
|
|
|
||
|
|
```bash
|
||
|
|
ollama create majestic-fin-r1 -f Modelfile
|
||
|
|
ollama run majestic-fin-r1
|
||
|
|
```
|
||
|
|
|
||
|
|
## Free hosted demo and API
|
||
|
|
|
||
|
|
A public Hugging Face Space can serve the `Q8_0` build on free CPU hardware. The companion Space for this repo is:
|
||
|
|
|
||
|
|
- `https://huggingface.co/spaces/EREN121232/MAJESTIC-FIN-R1-Free-API`
|
||
|
|
|
||
|
|
Once the Space is live, use the footer link `Use via API` to inspect endpoints, or call the `/chat` endpoint directly from Python, JavaScript, or curl.
|