Instructions to use SerialKicked/Lethe-AI-Repo with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use SerialKicked/Lethe-AI-Repo with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="SerialKicked/Lethe-AI-Repo",
	filename="emotion-bert-classifier.gguf",
)

output = llm(
	"Once upon a time,",
	max_tokens=512,
	echo=True
)
print(output)

Notebooks
Google Colab
Kaggle
Local Apps

llama.cpp

How to use SerialKicked/Lethe-AI-Repo with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf SerialKicked/Lethe-AI-Repo:Q6_K
# Run inference directly in the terminal:
llama-cli -hf SerialKicked/Lethe-AI-Repo:Q6_K

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf SerialKicked/Lethe-AI-Repo:Q6_K
# Run inference directly in the terminal:
llama-cli -hf SerialKicked/Lethe-AI-Repo:Q6_K

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf SerialKicked/Lethe-AI-Repo:Q6_K
# Run inference directly in the terminal:
./llama-cli -hf SerialKicked/Lethe-AI-Repo:Q6_K

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf SerialKicked/Lethe-AI-Repo:Q6_K
# Run inference directly in the terminal:
./build/bin/llama-cli -hf SerialKicked/Lethe-AI-Repo:Q6_K

Use Docker

docker model run hf.co/SerialKicked/Lethe-AI-Repo:Q6_K

LM Studio
Jan
Ollama
How to use SerialKicked/Lethe-AI-Repo with Ollama:
```
ollama run hf.co/SerialKicked/Lethe-AI-Repo:Q6_K
```

Unsloth Studio new

How to use SerialKicked/Lethe-AI-Repo with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for SerialKicked/Lethe-AI-Repo to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for SerialKicked/Lethe-AI-Repo to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for SerialKicked/Lethe-AI-Repo to start chatting

Docker Model Runner
How to use SerialKicked/Lethe-AI-Repo with Docker Model Runner:
```
docker model run hf.co/SerialKicked/Lethe-AI-Repo:Q6_K
```

Lemonade

How to use SerialKicked/Lethe-AI-Repo with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull SerialKicked/Lethe-AI-Repo:Q6_K

Run and chat with the model

lemonade run user.Lethe-AI-Repo-Q6_K

List all available models

lemonade list

SerialKicked commited on Mar 20

Commit

d804a91

verified ·

1 Parent(s): ad559a7

Update README.md

Browse files

Files changed (1) hide show

README.md +1 -1

README.md CHANGED Viewed

@@ -11,7 +11,7 @@ This repo contains useful content for the [Lethe AI Sharp](https://github.com/Se
 ## Fixed Jinja Templates
-This repo also contains fixed (and more permissive) Jinja templates for Mistral (Tekken7) and Qwen 3.5 (ChatML) models. They allow for system messages mid conversations (requirement for LetheAI) while other so-called errors that could trigger when using those LLM in way not initially intended.
 ## gte-large.Q6_K.gguf

 ## Fixed Jinja Templates
+This repo also contains fixed (and more permissive) Jinja templates for Mistral (Tekken7) and Qwen 3.5 (ChatML) models. They allow for system messages mid conversations (requirement for LetheAI), and fixes other so-called errors that could trigger when using those LLM in way not initially intended.
 ## gte-large.Q6_K.gguf