Ecosyste.ms: Issues
An open API service for providing issue and pull request metadata for open source projects.
GitHub / huggingface/safetensors issues and pull requests
#561 - Feature Request: Support for Ellipsis (...) in Indexing
Issue -
State: open - Opened by csaybar 4 days ago
#560 - Thread panicked in slice.rs due to out-of-range index in safetensors
Issue -
State: open - Opened by csaybar 4 days ago
#559 - serialize & deserialize does not work as the documentation specify.
Issue -
State: open - Opened by csaybar 6 days ago
#558 - Updating the dev number.
Pull Request -
State: closed - Opened by Narsil 10 days ago
#557 - Fix wrong signature of `safe_open.__init__` in stub file
Pull Request -
State: closed - Opened by SunghwanShim 11 days ago
- 2 comments
#556 - support no_std
Pull Request -
State: closed - Opened by ivila 15 days ago
#555 - Reducing the amount of copies during save.
Pull Request -
State: closed - Opened by Narsil 16 days ago
- 2 comments
#554 - Fixing the stub?
Pull Request -
State: closed - Opened by Narsil 16 days ago
- 1 comment
#553 - Memory map as private pure copy.
Pull Request -
State: closed - Opened by Narsil 16 days ago
#552 - Revert "Upgrade version number. (#551)"
Pull Request -
State: closed - Opened by Narsil 16 days ago
#551 - Upgrade version number.
Pull Request -
State: closed - Opened by Narsil 16 days ago
#550 - Upgrade version number.
Pull Request -
State: closed - Opened by Narsil 16 days ago
#549 - After resaving safetensors, there is no metadata.get("format") method
Issue -
State: closed - Opened by CallmeZhangChenchen 19 days ago
- 1 comment
#548 - Compatibility Issue with Python 3.13
Issue -
State: closed - Opened by pocerberus 20 days ago
- 1 comment
#547 - Cannot install safetensors: maturin failed project.version field is required in pyproject.toml
Issue -
State: closed - Opened by johnpc 22 days ago
- 4 comments
#545 - Use `mmap.MAP_PRIVATE` flag for loading .safetensors files on filesystems that don't support `SHARED`
Issue -
State: closed - Opened by koutini-sony about 1 month ago
#544 - Support no_std
Issue -
State: closed - Opened by ivila about 2 months ago
- 2 comments
#543 - Upgrading PyO3 to 0.23
Pull Request -
State: closed - Opened by IvanIsCoding about 2 months ago
- 1 comment
#542 - Don't store unused tensors in memory when loading a safetensors file
Issue -
State: closed - Opened by Nerogar 2 months ago
- 4 comments
#541 - [Question] Safetensors seem to block the main thread -- but torch.save does not?
Issue -
State: open - Opened by vedantroy 2 months ago
- 1 comment
#540 - Attempt to fix bench + update TF + maturin abi3 CI.
Pull Request -
State: closed - Opened by Narsil 2 months ago
#539 - Trying some new CI.
Pull Request -
State: closed - Opened by Narsil 2 months ago
#538 - Improve typing annotations for public apis
Issue -
State: open - Opened by justinchuby 2 months ago
- 1 comment
#537 - Image in huggingface docs has typo
Issue -
State: open - Opened by andrewkchan 3 months ago
- 3 comments
#536 - cannot import name 'save_open' from 'safetensors'
Issue -
State: closed - Opened by sun2ot 3 months ago
- 2 comments
#535 - Add support for Cambricon mlu devices
Pull Request -
State: closed - Opened by huismiling 3 months ago
- 8 comments
#534 - Update python-bench.yml
Pull Request -
State: closed - Opened by RoyelSteven-Ai 3 months ago
- 1 comment
#533 - Update __init__.pyi
Pull Request -
State: closed - Opened by w1gs 3 months ago
#532 - Documentation about multipart safetensors
Issue -
State: closed - Opened by attashe 3 months ago
- 6 comments
#531 - Can't save checkpoint to file descriptor.
Issue -
State: closed - Opened by GLivshits 4 months ago
- 1 comment
#530 - fix(save_file): add force_contiguous(default=True) to save_file
Pull Request -
State: closed - Opened by dbyoung18 4 months ago
- 4 comments
#529 - let `safetensors.load` support load weight from bytearray
Issue -
State: open - Opened by JeryZeng 4 months ago
- 1 comment
#528 - RuntimeError: unable to mmap 23802932552 bytes from file <./flux1-dev.safetensors>: Cannot allocate memory (12)
Issue -
State: open - Opened by sakura-nyaa 4 months ago
- 1 comment
#527 - [Question] Comparison with the zarr format?
Issue -
State: open - Opened by julioasotodv 4 months ago
- 13 comments
#526 - Improve clarity in safetensors' `index.mdx` doc
Pull Request -
State: closed - Opened by ethanknights 4 months ago
- 1 comment
#525 - Prepare for 0.4.5
Pull Request -
State: closed - Opened by Narsil 4 months ago
#524 - add remapping to safetensors.torch.load_model
Issue -
State: closed - Opened by f-fuchs 5 months ago
- 1 comment
#523 - SafetensorError does not appear in `__init__.pyi`
Issue -
State: closed - Opened by alanhdu 5 months ago
- 1 comment
#521 - Retrieving saved metadata
Issue -
State: closed - Opened by parlance-zz 5 months ago
- 3 comments
#520 - Some layers are not loaded(‘encoder.embed_token.weight’ and 'shared.weight') when loading converted safetensors into transfomrers.MT5EncoderModel.
Issue -
State: open - Opened by WongGawa 5 months ago
- 1 comment
#519 - converting weight of more than 50G‘s .bin to .safetensors failed
Issue -
State: open - Opened by WongGawa 5 months ago
- 1 comment
#518 - fix lifetimes, add tensors iterator
Pull Request -
State: closed - Opened by gvilums 5 months ago
- 3 comments
#517 - Use id rather than modelId
Pull Request -
State: closed - Opened by osanseviero 5 months ago
#516 - Adding support for wrapper tensor subclass for safetensor
Pull Request -
State: closed - Opened by jerryzh168 5 months ago
- 6 comments
#515 - Problem serializing quantized weights
Issue -
State: closed - Opened by sayakpaul 5 months ago
- 3 comments
#514 - Forcing empty list/empty tuple behavior.
Pull Request -
State: closed - Opened by Narsil 6 months ago
#513 - Fixing release workflow.
Pull Request -
State: closed - Opened by Narsil 6 months ago
#512 - Git v0.4.4 rc0
Pull Request -
State: closed - Opened by Narsil 6 months ago
Labels: Full Build
#511 - Upgrading the version.
Pull Request -
State: closed - Opened by Narsil 6 months ago
#510 - Fixing zero-width tensor for in memory loading.
Pull Request -
State: closed - Opened by Narsil 6 months ago
- 1 comment
#509 - Respects torch.device(0) new behavior without breaking backward compatibility
Pull Request -
State: closed - Opened by Narsil 6 months ago
- 2 comments
#508 - Change doc-builder workflow container
Pull Request -
State: closed - Opened by mishig25 6 months ago
- 2 comments
#507 - Improving bf16 tests
Pull Request -
State: closed - Opened by Narsil 6 months ago
- 2 comments
#506 - Revert "Improving the bf16 tests for PT+TF."
Pull Request -
State: closed - Opened by Narsil 6 months ago
#505 - Improving the bf16 tests for PT+TF.
Pull Request -
State: closed - Opened by Narsil 6 months ago
#504 - Upgrading Pyo3 0.22 (with gil-refs)
Pull Request -
State: closed - Opened by Narsil 6 months ago
#503 - Wrong device when using device="cpu" with torch.device
Issue -
State: open - Opened by chrisdesa 6 months ago
- 4 comments
#501 - Add `safetensors.torch.save_state_dict()` to python bindings
Pull Request -
State: closed - Opened by mohahf19 6 months ago
- 1 comment
#500 - Query device name from pytorch if only device index is given
Pull Request -
State: closed - Opened by dvrogozh 6 months ago
- 11 comments
#499 - pytorch: safetensors library hardcodes using CUDA if only device index is provided
Issue -
State: closed - Opened by dvrogozh 6 months ago
- 3 comments
#498 - get_tensor returns an empty tensor on LoRa safetensors adapter saved by SFTTrainer
Issue -
State: closed - Opened by tsvisab 6 months ago
- 1 comment
#497 - Add conversion validation support for the float8 tensor dtype
Pull Request -
State: closed - Opened by darthhexx 6 months ago
- 1 comment
#496 - get_slice has the same speed as get_tensor
Issue -
State: closed - Opened by youkaichao 6 months ago
- 1 comment
#495 - inconsistent behavior of slicing in certain model file
Issue -
State: closed - Opened by youkaichao 6 months ago
- 4 comments
#494 - Add support for xla devices
Pull Request -
State: closed - Opened by BitPhinix 7 months ago
- 1 comment
#492 - SafetensorError is not pickleable, potentially causing hangs in multi-process environments
Issue -
State: closed - Opened by Wehrdo 7 months ago
- 2 comments
Labels: Stale
#491 - Compilation fail due to test error
Issue -
State: closed - Opened by BananPrzydawka 7 months ago
- 3 comments
#490 - How to save model checkpoint from a distributed training from multiple nodes?
Issue -
State: closed - Opened by Emerald01 7 months ago
- 2 comments
Labels: Stale
#489 - Unable to save llama3 since complex64 is not supported
Issue -
State: closed - Opened by npuichigo 7 months ago
- 1 comment
Labels: Stale
#488 - Minor correction about Arrow in README.md
Pull Request -
State: closed - Opened by lhoestq 7 months ago
- 1 comment
Labels: Stale
#487 - README is inaccurate
Issue -
State: closed - Opened by judahrand 7 months ago
- 3 comments
Labels: Stale
#486 - Register a MIME type for the `safetensors` format.
Issue -
State: closed - Opened by judahrand 7 months ago
- 1 comment
Labels: Stale
#485 - misleading error message in `safe_open`
Issue -
State: closed - Opened by Clement-Lelievre 7 months ago
- 5 comments
Labels: Stale
#484 - feat(ci): add trufflehog secrets detection
Pull Request -
State: closed - Opened by McPatate 7 months ago
#482 - ModuleNotFoundError: No module named 'torch._higher_order_ops'
Issue -
State: closed - Opened by guotong1988 8 months ago
- 1 comment
#481 - Error when using map_location="cuda:0" with multiple GPUs
Issue -
State: closed - Opened by AAnirudh07 8 months ago
- 1 comment
Labels: Stale
#480 - safetensors.tensorflow.save modifies its input argument
Issue -
State: closed - Opened by allaisandrea 8 months ago
- 1 comment
Labels: Stale
#479 - support for Cambricon MLUs device
Pull Request -
State: closed - Opened by huismiling 8 months ago
- 1 comment
Labels: Stale
#478 - Can't seem to skip parameter initialization while using the `safetensors.torch.load_model` API!
Issue -
State: closed - Opened by goelayu 8 months ago
- 1 comment
Labels: Stale
#477 - Support multi-device loading for load_file() with pt framework
Issue -
State: closed - Opened by mobicham 8 months ago
- 2 comments
Labels: Stale
#476 - Inaccurate performance benchmarks
Issue -
State: closed - Opened by goelayu 9 months ago
- 1 comment
Labels: Stale
#475 - Official C++ API lib/API
Issue -
State: closed - Opened by dharmeshkakadia 9 months ago
- 1 comment
Labels: Stale
#474 - How to fully load checkpointed weights in memory?
Issue -
State: closed - Opened by goelayu 9 months ago
- 1 comment
#472 - Fixing empty serialization (no tensor) with some metadata.
Pull Request -
State: closed - Opened by Narsil 9 months ago
#471 - Request for writable temporary directory handling within Python tests
Issue -
State: closed - Opened by salayatana66 9 months ago
- 1 comment
Labels: Stale
#470 - Add `Cargo.lock` (again)
Pull Request -
State: closed - Opened by 9999years 9 months ago
Labels: Stale
#469 - error saving AI model weights with non-contiguous tensors
Issue -
State: closed - Opened by not-lain 9 months ago
- 1 comment
Labels: Stale
#468 - Upgrading pyo3 to 0.21.
Pull Request -
State: closed - Opened by Narsil 9 months ago
#467 - Fix incorrect serialization given only metadata
Pull Request -
State: closed - Opened by tommyip 9 months ago
Labels: Stale
#466 - Incorrect serialization given only metadata
Issue -
State: closed - Opened by tommyip 9 months ago
- 1 comment
Labels: Stale
#465 - Updating doc.
Pull Request -
State: closed - Opened by Narsil 9 months ago
- 1 comment
#463 - Allow serialization of TensorInfo, and other safetensors classes
Issue -
State: closed - Opened by LeoDog896 10 months ago
- 7 comments
#462 - docs: fix all typos
Pull Request -
State: closed - Opened by LeoDog896 10 months ago
Labels: Stale
#461 - docs: fix typo Boolan -> Boolean
Pull Request -
State: closed - Opened by LeoDog896 10 months ago
- 1 comment
#460 - Poor performance for reading Numpy
Issue -
State: closed - Opened by ZHUI 10 months ago
- 11 comments
Labels: Stale
#459 - Typing error using with safe_open
Issue -
State: closed - Opened by chanind 10 months ago
- 1 comment
Labels: Stale
#458 - Safetensors uses excessive RAM when saving files
Issue -
State: closed - Opened by sheepymeh 10 months ago
- 1 comment
Labels: Stale
#457 - Extend data type support for MLX's metadata
Issue -
State: closed - Opened by mzbac 10 months ago
- 1 comment
Labels: Stale
#456 - Update convert.py to allow local (default) cache to be used
Pull Request -
State: closed - Opened by sealad886 10 months ago
Labels: Stale