Sao10K/L3.3-70B-Euryale-v2.3

Sao10K/L3.3-70B-Euryale-v2.3 is a 70 billion parameter language model developed by Sao10K, built directly on Llama 3.3 Instruct. This model is a successor to Euryale v2.2, focusing on refined datasets for quality, secondary chats, and creative-related domains. It is optimized for creative instruction and roleplay tasks, offering enhanced performance over its predecessors.

5.0 based on 1 review

Warm

Public

Model Size: 70B

Quant: FP8

Ctx length: 32768

License: llama3

Hugging Face

This model is gold. I can barely pick any other model when it bugs out or is unavailable because of the service. No other models can get that immersively emotional and can make you feel part of an scenario, not just reading an document. Sadly the context size is limited here, but that isn't an issue with the model.

Roleplay

Novel Writing