gguf-py : add IQ1_M to GGML_QUANT_SIZES (#6761)

pmysl · web-flow · commit c1386c936e9f · 2024-04-21T15:49:30.000+03:00
diff --git a/gguf-py/gguf/constants.py b/gguf-py/gguf/constants.py
@@ -872,6 +872,7 @@ def get_type(val: Any) -> GGUFValueType:
     GGMLQuantizationType.I32:     (1, 4),
     GGMLQuantizationType.I64:     (1, 8),
     GGMLQuantizationType.F64:     (1, 8),
+    GGMLQuantizationType.IQ1_M:   (256, QK_K // 8 + QK_K // 16  + QK_K // 32),
 }
 
 

Original file line number	Diff line number	Diff line change
`@@ -872,6 +872,7 @@ def get_type(val: Any) -> GGUFValueType:`
`872`	`872`	`GGMLQuantizationType.I32: (1, 4),`
`873`	`873`	`GGMLQuantizationType.I64: (1, 8),`
`874`	`874`	`GGMLQuantizationType.F64: (1, 8),`
	`875`	`+ GGMLQuantizationType.IQ1_M: (256, QK_K // 8 + QK_K // 16 + QK_K // 32),`
`875`	`876`	`}`
`876`	`877`
`877`	`878`