Text Generation
Safetensors
qwen3_5
conversational
Eval Results

Feedback

#1
by KeyboardMasher - opened

I've tested this model in IQ4_XS quantization by bartowski. Unfortunately I am seeing a lot of hallucinations in the result output both on factual recall and reasoning prompts. Hope you can improve that in future releases.

Cheers!

Orion LLM Labs org

Hello! Thanks for the feedback! I dont recomend IQ4_XS quantization because its bad, i recomend using the full model or use a better quantization, or a highter quantization

Sign up or log in to comment