FOXAI SEARCH
전체 검색
AI 모델 85012
- gemma-4-26B-A4B-it-qat-assistant-q4_0-gguf↗
RachidAR
`Q4_0-q4emb` is an experimental quantization where `token_embd.weight` is kept in `Q4_0` instead of `Q6_K` precision quantization typically used by llama.cpp. This follows a similar approach to recent QAT experiments for Gemma models, where preserving the original Q4-trained embedding format may better match the intended QAT behavior. Initial testing showed similar draft acceptance rates to the default Q4_0 quant, with a small speed advantage, though more benchmarking is needed.
- Rain-100M↗
raincandy-u
- Laya-Multilingual-MXFP8↗
sahilchachra
- LFM3-1.7B-Paretrix↗
Soulfate24
- MedVAE↗
stanfordmimi
- Step-Audio-EditX↗
阶跃星辰
- py39-pylingual-v1-segmenter↗
syssec-utd
- Qwen3.6-35B-A3B-VQ-3.4bpw↗
TheDrainFlorist
- SkyReels-V2-I2V-14B-540P-GGUF↗
ThePun1sher
- convnext_xlarge.fb_in22k_ft_in1k_384↗
timm
