EchoLabs33/qwen2.5-14b-instruct-hxq
Text Generation • 15B • Updated • 194
HXQ Compressed Models
Calibration-free VQ-256 compression. Transformers, SSMs, hybrids. Beats GPTQ/AWQ quality. Zero calibration data.
-
-
EchoLabs33/qwen2.5-7b-instruct-hxq
Text Generation • 8B • Updated • 50 -
EchoLabs33/qwen2.5-3b-instruct-hxq
Text Generation • 3B • Updated • 45 -
EchoLabs33/qwen2.5-coder-3b-hxq
Text Generation • 3B • Updated • 114 -
EchoLabs33/zamba2-2.7b-instruct-hxq
Text Generation • 3B • Updated • 322 -
EchoLabs33/zamba2-1.2b-hxq
Text Generation • 1B • Updated • 14 -
EchoLabs33/tinyllama-1.1b-hxq
Text Generation • 1B • Updated • 7 -
EchoLabs33/mamba2-1.3b-hxq
Text Generation • 1B • Updated • 20 -
EchoLabs33/mamba-130m-hxq
Text Generation • 0.1B • Updated • 24