Conversation
Switch all weight and optimizer state serialization from JSON (f64 text) to safetensors (f32 binary), reducing checkpoint size ~5x and enabling f16 inference exports at ~10x reduction. - Add save_weights_st/load_weights_st to TensorGpt - Add save_m_st/save_v_st/load_state_st to TensorAdam - Add InferenceGpt::load_safetensors (handles f32 and f16) - Add serialize_state_dict_st for f16 export - Add `microgpt export --half` subcommand - Update CLI train/infer/info, serve to use safetensors - Remove all JSON weight serialization - Update all tests to safetensors; add byte-identity and f16 export tests - Update READMEs for lib, CLI, and serve Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Co-Authored-By: Claude Opus 4.6 (1M context) <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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Replace JSON weight format with safetensors across microgpt
Switch all weight and optimizer state serialization from JSON (f64 text)
to safetensors (f32 binary), reducing checkpoint size ~5x and enabling
f16 inference exports at ~10x reduction.
microgpt export --halfsubcommandCo-Authored-By: Claude Opus 4.6 (1M context) noreply@anthropic.com