Quantized Models (GGUF, IQ, Imatrix)
Collection
Various GGUF quantizations of small models. Models with a "checkmark" are personal favorites. An "orange arrow" means it's being uploaded. • 97 items • Updated • 72
How to use from
llama.cppInstall from WinGet (Windows)
winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf Lewdiculous/Aurora_l3_8B-GGUF-IQ-Imatrix:# Run inference directly in the terminal:
llama cli -hf Lewdiculous/Aurora_l3_8B-GGUF-IQ-Imatrix: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 Lewdiculous/Aurora_l3_8B-GGUF-IQ-Imatrix:# Run inference directly in the terminal:
./llama-cli -hf Lewdiculous/Aurora_l3_8B-GGUF-IQ-Imatrix: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 Lewdiculous/Aurora_l3_8B-GGUF-IQ-Imatrix:# Run inference directly in the terminal:
./build/bin/llama-cli -hf Lewdiculous/Aurora_l3_8B-GGUF-IQ-Imatrix:Use Docker
docker model run hf.co/Lewdiculous/Aurora_l3_8B-GGUF-IQ-Imatrix:Quick Links
Outdated:
Outdaded tokenizer configuration!
This is only kept for historical purposes, use the newer models instead of this one. Busy times, for the faint of heart, wait the storm.
This is yet another attempt with hopefully more stable formatting for quotes/asterisks dialogues.
GGUF-IQ-Imatrix quants for ResplendentAI/Aurora_l3_8B.
Recommended presets here or here.
Use the latest version of KoboldCpp. Use the provided presets.
This is all still highly experimental, modified configs were used to avoid the tokenizer issues, let the authors know how it performs for you, feedback is more important than ever now.
Original model information:
Aurora
A more poetic offering with a focus on perfecting the quote/asterisk RP format. I have strengthened the creative writing training.
Make sure your example messages and introduction are formatted cirrectly. You must respond in quotes if you want the bot to follow. Thoroughly tested and did not see a single issue. The model can still do plaintext/aserisks if you choose.
- Downloads last month
- 78
Hardware compatibility
Log In to add your hardware
3-bit
4-bit
5-bit
6-bit
8-bit
16-bit
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support
Install (macOS, Linux)
# Start a local OpenAI-compatible server with a web UI: llama serve -hf Lewdiculous/Aurora_l3_8B-GGUF-IQ-Imatrix:# Run inference directly in the terminal: llama cli -hf Lewdiculous/Aurora_l3_8B-GGUF-IQ-Imatrix: