Calibration-free VQ-256 compression. Transformers, SSMs, hybrids. Beats GPTQ/AWQ quality. Zero calibration data.
EchoLabs
non-profit
AI & ML interests
AI cross compression architecture
Recent Activity
View all activity
models 21
EchoLabs33/sentinel-qwen2.5-3b-security
Updated
EchoLabs33/zamba2-2.7b-instruct-hxq
Text Generation • 3B • Updated • 274
EchoLabs33/qwen2.5-7b-instruct-hxq
Text Generation • 8B • Updated • 177
EchoLabs33/qwen2.5-coder-3b-hxq
Text Generation • 3B • Updated • 389
EchoLabs33/qwen2.5-sentinel-3b-hxq
Updated
EchoLabs33/qwen2.5-3b-instruct-hxq
Text Generation • 3B • Updated • 159
EchoLabs33/qwen2.5-14b-instruct-hxq
Text Generation • 15B • Updated • 240
EchoLabs33/smollm3-3b-hxq
Text Generation • 3B • Updated • 8
EchoLabs33/granite-4.0-h-micro-hxq
Text Generation • 3B • Updated • 81
EchoLabs33/clip-vit-large-patch14-hxq
Zero-Shot Image Classification • 0.4B • Updated • 5