Datasets and model formats
Load and save datasets and model files in the formats you need.
Start here
The Featured projects with the most downloads.
-
arrow-rs
Rust implementation of Apache Arrow
11M downloads (30 days) Latest release: 60.1.0, 28 Oct 2026
-
safetensors
Provides functions to read and write safetensors which aim to be safer than their PyTorch counterpart.
5.5M downloads (30 days) Latest release: 0.8.0, 9 Jun 2026
-
hf-hub
Rust client for the Hugging Face Hub API
3.4M downloads (30 days) Latest release: 1.0.0, 10 Jul 2026
All projects
Order: strongest tier first, then most downloads, then name, with unmeasured values last. A library counts downloads over 30 days. An app counts binary downloads per release. An app with none counts its crate downloads over 30 days.
What do the tiers mean?
- Featured
- Shown first. It passes every Featured gate.
- Listed
- Shown in the normal listing. No other tier applies.
- New
- A project whose first release is recent.
- Archived
- No longer maintained, or ended by its author.
Tiers come from published rules over raw signals. Stars alone never qualify a project.
Active dependents: crates from other owners that depend on this project in a current release. One owner counts at most once.
Movement needs more history.
| Project | Downloads | Dependents | Stars | Latest release | Runs on (GPU) | Native code | License |
|---|---|---|---|---|---|---|---|
| Featured 13 projects | |||||||
|
arrow-rs Featured
Rust implementation of Apache Arrow Python bindings |
10,746,755 downloads over 30 days | 2,126 | 3,664 | 60.1.0 | GPU support unknown | Binds native code: ring (feature encryption), liblzma-sys (feature default), zstd-sys (feature default) | Apache-2.0 |
|
safetensors Featured
Provides functions to read and write safetensors which aim to be safer than their PyTorch counterpart. |
5,486,837 downloads over 30 days | 531 | 3,943 | 0.8.0 | GPU support unknown | No native code found | Apache-2.0 |
|
hf-hub Featured
Rust client for the Hugging Face Hub API |
3,442,405 downloads over 30 days | 578 | 343 | 1.0.0 | GPU support unknown | No native code found | Apache-2.0 |
|
datafusion Featured
DataFusion is an in-memory query engine that uses Apache Arrow as the memory model |
2,992,524 downloads over 30 days | 484 | 9,578 | 55.2.0 | GPU support unknown | Binds native code: libmimalloc-sys, liblzma-sys (feature compression), zstd-sys (feature compression) | Apache-2.0 |
|
polars Featured
DataFrame library based on Apache Arrow Python bindings |
1,729,523 downloads over 30 days | 639 | 40,285 | 0.55.2 | GPU support unknown | Binds native code: zstd-sys (feature decompress) | MIT |
|
lance Featured
A columnar data format that is 100x faster than Parquet for random access. |
915,526 downloads over 30 days | 48 | 7,268 | 14.0.2 | GPU support unknown | Binds native code: ring (feature credential-vendor-gcp), lz4-sys (feature lz4), zstd-sys (feature zstd). Build uses a native toolchain: cc | Apache-2.0 |
|
ndarray-npy Featured
.npy and .npz file format support for ndarray |
560,950 downloads over 30 days | 99 | 71 | 0.10.0 | GPU support unknown | No native code found | Apache-2.0 OR MIT |
|
npyz Featured
NumPy file format (de-)serialization. Fork of outdated npy-rs. |
467,634 downloads over 30 days | 36 | 33 | 0.9.1 | GPU support unknown | No native code found | MIT |
|
delta-rs Featured
Native Delta Lake implementation in Rust |
463,085 downloads over 30 days | 33 | 3,356 | 1.1.4 | GPU support unknown | Binds native code: lz4-sys (feature delta-cache), zstd-sys (feature delta-cache) | Apache-2.0 |
|
hdf5-rust Featured
Thread-safe Rust bindings for the HDF5 library. |
253,180 downloads over 30 days | 70 | 85 | 0.15.1 | GPU support unknown | Binds native code: hdf5-metno-sys, blosc-src (feature blosc), lzf-sys (feature lzf), mpi-sys (feature mpio), zfp-sys (feature zfp) | Apache-2.0 OR MIT |
|
zarrs Featured
A library for the Zarr storage format for multidimensional arrays and metadata |
75,052 downloads over 30 days | 51 | 285 | 0.23.14 | GPU support unknown | Binds native code: blosc-src (feature blosc), gdeflate-sys (feature gdeflate), zfp-sys (feature zfp), zstd-sys (feature zstd) | Apache-2.0 OR MIT |
|
hdf5-pure Featured
Pure-Rust HDF5 library: read, write, and edit files in place (WASM-compatible, no C dependencies) |
64,279 downloads over 30 days | 18 | 42 | 0.47.0 | GPU support unknown | No native code found | Apache-2.0 OR MIT |
|
netcdf Featured
High-level NetCDF bindings for Rust |
22,481 downloads over 30 days | 29 | 106 | 0.12.1 | GPU support unknown | Binds native code: netcdf-sys, hdf5-metno-sys, libz-sys, mpi-sys (feature mpi) | Apache-2.0 OR MIT |
| Listed 20 projects | |||||||
|
xet-core Listed
Async runtime, configuration, logging, and utility infrastructure for the Hugging Face Xet client tools. Python bindings |
467,103 downloads over 30 days | 9 | 625 | 1.8.0 | GPU support unknown | No native code found | Apache-2.0 |
|
vortex Listed
An extensible, state-of-the-art framework for columnar compression, and the fastest FOSS columnar file format. Formerly at @spiraldb, now an Incubation Stage project at LFAI&Data, part of the Linux Foundation. |
98,129 downloads over 30 days | 32 | 3,310 | 0.93.0 | GPU support unknown | Binds native code: zstd-sys | Apache-2.0 |
|
dlpark Listed
Pure Rust DLPack implementation for zero-copy tensor interoperability Python bindings |
68,157 downloads over 30 days | 5 | 66 | 0.10.0 | CUDA | No native code found | Apache-2.0 |
|
netcdf-rust Listed
Pure-Rust NetCDF-3 classic and NetCDF-4 (HDF5-backed) file reader |
36,012 downloads over 30 days | 10 | 11 | 0.12.1 | GPU support unknown | No native code found | Apache-2.0 OR MIT |
|
ndarray-csv Listed
Easily read and write homogeneous CSV data to and from 2D ndarrays |
23,808 downloads over 30 days | 7 | 20 | 0.5.4 | GPU support unknown | No native code found | Apache-2.0 OR MIT |
|
burn-onnx Listed
Import and export ONNX models with the Burn framework |
14,530 downloads over 30 days | 23 | 141 | 0.22.0 | GPU support unknown | No native code found | Apache-2.0 OR MIT |
|
Matfile is a library for reading and writing Matlab ".mat" data files. |
10,737 downloads over 30 days | 5 | 14 | 0.5.0 | GPU support unknown | No native code found | MIT |
|
MNIST data set parser. |
7,083 downloads over 30 days | 1 | 23 | 0.6.0 | GPU support unknown | Binds native code: curl-sys (feature download), libz-sys (feature download) | MIT |
|
connector-x Listed
Load data from databases to dataframes, the fastest way. |
4,951 downloads over 30 days | 5 | 2,659 | 0.4.6 | GPU support unknown | Binds native code: libsqlite3-sys (feature src_sqlite), odpic-sys (feature src_oracle), openssl-sys (feature src_postgres) | MIT |
|
onnx-extractor Listed
Minimal ONNX model loader for extracting weights, tensors, operations, and graph structure |
3,975 downloads over 30 days | 4 | 6 | 0.5.2 | GPU support unknown | No native code found | MIT |
|
gguf Listed
GGUF file parser |
2,472 downloads over 30 days | 4 | 25 | 0.1.8 | GPU support unknown | No native code found | MIT |
|
onnx-rs Listed
Zero-dependency ONNX model parser in Rust |
453 downloads over 30 days | 5 | 6 | 0.1.3 | GPU support unknown | No native code found | MIT |
|
Oxen server is a fast data version control backend, supporting local disk and S3. Self host your repositories on your own storage, or use the hosted platform on Oxen.ai. Stores, syncs, and serves versioned datasets, model checkpoints, game assets, studio media, and any large data. |
126 binary downloads per release | Not applicable | 1,208 | 0.67.0 | GPU support unknown | Binds native code: librocksdb-sys, lmdb-master-sys, ffmpeg-sys-next (feature ffmpeg), libduckdb-sys (feature docs) | Apache-2.0 |
|
Data Agent Ready Warehouse : One for Analytics, Search, AI, Python Sandbox. — rebuilt from scratch. Unified architecture on your S3. |
48 binary downloads per release | Not applicable | 9,493 | Not measured | Not measured | Apache-2.0 | |
|
🔨 🍇 💻 🚀 GraphScope: A One-Stop Large-Scale Graph Computing System from Alibaba | 一站式图计算系统 |
48 binary downloads per release | Not applicable | 3,563 | Not measured | Not measured | Apache-2.0 | |
|
elusion Listed
Elusion is a modern DataFrame / Data Engineering / Data Analysis library that combines the familiarity of DataFrame operations (like those in PySpark, Pandas, and Polars) with the power of SQL query building. |
27 downloads over 30 days | 0 | 144 | 9.0.0 | GPU support unknown | Binds native code: ring, openssl-sys | MIT |
|
One backbone, many specialists. CMF is a self-describing, memory-mappable single-file format and a dependency-free runtime for quantized LLMs — zero-copy execution on CPU or GPU (Vulkan · Metal · DX12) that overlays task-specialized skills onto one shared model, with no per-model duplication. |
24 binary downloads per release | Not applicable | 52 | Not measured | Not measured | Apache-2.0 | |
|
The pipeline for the OSCAR corpus. |
1 downloads over 30 days | 0 | 178 | 2.0.0 | GPU support unknown | Binds native code: zstd-sys | Apache-2.0 |
|
Drop-in Apache Spark replacement written in Rust, unifying batch processing, stream processing, and compute-intensive AI workloads. |
0 binary downloads per release | Not applicable | 3,491 | Not measured | Not measured | Apache-2.0 | |
|
A Package Manager for Machine Learning Datasets and Models. |
Not measured | Not applicable | 85 | Not measured | Not measured | Not measured | BSD-3-Clause |
| New 2 projects | |||||||
|
hdf5-pure-rs New
Pure Rust implementation of the HDF5 file format |
801 downloads over 30 days | 5 | 5 | 0.3.10 | GPU support unknown | No native code found | BSD-3-Clause |
|
Model conversion and quantization with no Python: reads SafeTensors, GGUF, TCF and PyTorch state dicts, writes SafeTensors, GGUF and TCF, and quantizes to Q2_K through Q8_0 and IQ4 formats, including Q4_K_M mixes. GPTQ, AWQ, NF4, FP8, EXL2 and ONNX input are not implemented. Python bindings |
23 downloads over 30 days | Not applicable | 0 | 0.2.1 | GPU support unknown | No native code found | Apache-2.0 |
| Archived 2 projects | |||||||
|
black-jack Archived
DataFrame / Series / general data processing in Rust |
1 downloads over 30 days | 0 | 31 | 0.1.0 | GPU support unknown | Binds native code: GSL-sys | GPL-3.0 |
|
A Rust DataFrame implementation, built on Apache Arrow |
Not measured | Not applicable | 280 | Not measured | Not measured | Not measured | Apache-2.0 |