fix: Add Float64 dtype support for safetensors conversion#309
Open
ymote wants to merge 1 commit intooxiglade:mainfrom
Open
fix: Add Float64 dtype support for safetensors conversion#309ymote wants to merge 1 commit intooxiglade:mainfrom
ymote wants to merge 1 commit intooxiglade:mainfrom
Conversation
Previously, attempting to convert an f64 Array to a safetensors TensorView would fall through to the Complex64 error case. This adds proper handling for Float64 dtype, enabling save/load of f64 arrays via safetensors. This is needed for loading certain model weights (e.g., some normalization layers) that use f64 precision. Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Float64(f64) dtype support in theTryFrom<&Array> for TensorViewimplementationProblem
Previously, attempting to convert an f64 MLX Array to a safetensors
TensorViewwould fall through to theComplex64error case, making it impossible to save f64 arrays to safetensors files.Solution
Add the missing
Dtype::Float64match arm that properly converts f64 arrays to safetensors format.Use Case
This is needed for loading/saving certain model weights (e.g., some normalization layers) that use f64 precision.
Test plan
🤖 Generated with Claude Code