daavidhauser
/

Swift-1.5-Qwen3.8-27B-W4A16-HyperQwen-INT4

Text Generation

compressed-tensors

efficient-thinking

Model card Files Files and versions

Swift-1.5-Qwen3.8-27B-W4A16-HyperQwen-INT4 / RUNTIME.md

Commit History

Rename Swift 1.5 fast variant to INT4 heads

64515d3
verified

daavidhauser commited on 11 days ago

Release Swift 1.5 HyperQwen fast quantization with RTX 3090 evaluation

1613e65
verified

daavidhauser commited on 11 days ago