mirror of
https://github.com/ollama/ollama.git
synced 2026-08-04 14:56:15 +00:00
964ea42c09
This is a rewrite of the create functionality for the MLX engine. The core idea behind the create functionality is to break the import/convert into a pipeline of distinct phases: * Read (scan the safetensors directory for the various bits of metadata) * Classify (determine what the import type) * Plan (determine any transforms that need to be done) * Write (transform any data as necessary and write out the blobs) * Create the manifest Each architecture has a "policy" which determines how to convert the model correctly. A number of different formats for safetensors are supported including: * nvfp4 (two formats: model optimized, torch) * fp8 datatypes (convert to mxfp8) * standard bf16 based weights A number of cleanups/simplifications have been done including: * using the baked in names for the tensors instead of munging them into something else * unified 3d expert tensors (instead of separate per expert tensors) * fewer unnecessary transforms to the various tensors in a model (keep a model as close to the source as possible) * unified capability checking * draft model handling (for MTP) is done on the same path Image generation has been intentionally removed.
23 lines
561 B
Go
23 lines
561 B
Go
package create
|
|
|
|
import (
|
|
"encoding/json"
|
|
"strings"
|
|
)
|
|
|
|
type lagunaImportTransform struct{}
|
|
|
|
func newLagunaImportTransform(json.RawMessage) (quantizePolicy, error) {
|
|
return lagunaImportTransform{}, nil
|
|
}
|
|
|
|
func (lagunaImportTransform) quantizationType(name string, shape []int32, quantize string) string {
|
|
if !lagunaIsHFRoutedExpertWeight(name) {
|
|
return ""
|
|
}
|
|
return GetTensorQuantization(name, shape, quantize)
|
|
}
|
|
|
|
func lagunaIsHFRoutedExpertWeight(name string) bool {
|
|
return strings.HasSuffix(name, ".weight") && strings.Contains(name, ".mlp.experts.")
|
|
}
|