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 • 73
This repository hosts GGUF-IQ-Imatrix quantizations for KatyTheCutie/LemonadeRP-4.5.3.
--contextsize 8192
What does "Imatrix" mean?
It stands for Importance Matrix, a technique used to improve the quality of quantized models. The Imatrix is calculated based on calibration data, and it helps determine the importance of different model activations during the quantization process. The idea is to preserve the most important information during quantization, which can help reduce the loss of model performance, especially when the calibration data is diverse. [1] [2]
For imatrix data generation, kalomaze's groups_merged.txt with added roleplay chats was used, you can find it here.
Steps:
Base⇢ GGUF(F16)⇢ Imatrix-Data(F16)⇢ GGUF(Imatrix-Quants)
Quants:
quantization_options = [
"Q4_K_M", "IQ4_XS", "Q5_K_M", "Q5_K_S", "Q6_K",
"Q8_0", "IQ3_M", "IQ3_S", "IQ3_XXS"
]
If you want anything that's not here or another model, feel free to request.
Original model information:
Lemonade RP 4.5.3
8192 context length.
7B roleplay focused model, creativity and less cliché is the focus of this merge.
SillyTavern settings:
Models used in merge:
- NeverSleep/Noromaid-7B-0.4-DPO
- cgato/Thespis-7b-v0.5-SFTTest-2Epoch
- NurtureAI/neural-chat-7b-v3-1-16k
- cgato/Thespis-CurtainCall-7b-v0.2.2
- tavtav/eros-7b-test
Feedback is always greatly appreciated! <3
- Downloads last month
- 413
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