Could you please update the vectionlabs/Salience-1.5-Flash GGUFs with the latest version?
#2647
by vectionlabs - opened
The current GGUF files are outdated. We noticed a high rate of model hallucinations when handling ambiguous user prompts, which has been successfully fixed in our latest model release. Could you please update the GGUF weights with the new version?
This has been addressed. The old quants got nuked, the model reconverted, requantized and the imatrix recomputed. All the new quants are now uploaded.
Convenient download page: https://hf.tst.eu/model#Salience-1.5-Flash-GGUF
Static quants: https://huggingface.co/mradermacher/Salience-1.5-Flash-GGUF
Weighted/imatrix quants: https://huggingface.co/mradermacher/Salience-1.5-Flash-i1-GGUF