← Package directory
Available on winget

Install HuggingFace Model Downloader

Simple go utility to download HuggingFace Models and Datasets

Install with winget
winget install --id bodaay.hfdownloader
Upgrade
winget upgrade --id bodaay.hfdownloader
Uninstall
winget uninstall --id bodaay.hfdownloader

About HuggingFace Model Downloader

The HuggingFace Model Downloader is a utility tool for downloading models and datasets from the HuggingFace website. It offers multithreaded downloading for LFS files and ensures the integrity of downloaded models with SHA256 checksum verification.

What's new in 3.0.4

What's Changed in v3.0.4 Bug-fix release tackling eight open issues reported against v3.0.3. Focused on download correctness, webui responsiveness, storage flexibility, and analyzer coverage for new quant formats. Every fix ships with regression tests; go test -race ./... is now fully green across every package. Bug Fixes - Resumable downloads actually resume now (#70) — downloadSingle and each downloadMultipart part goroutine now open .part files with O_RDWR|O_CREATE (no truncate), stat the existing size, and issue Range: bytes=pos-end requests from that offset. Interrupted downloads resume correctly on rerun, and within-process retries after a flaky-connection cut no longer lose bytes. Previously every Ctrl+C silently re-fetched from zero and single-file downloads literally had no resume code path at all. - Unsloth Dynamic quant variants correctly labeled (#72) — the GGUF quant regex now supports Q{2..8}_K_XL / _XXL suffixes and IQ1_S / IQ1_M. Previously on repos like unsloth/Qwen3-30B-A3B-GGUF the regex silently collapsed six UD-Q*_K_XL files into Q2_K..Q8_K labels that collided with the real plain-K quants, and missed the two IQ1 files entirely. All 25 GGUF files in that repo now label correctly. Added quality/description map entries for Q2_K_L, Q2..8_K_XL, IQ1_S/M, IQ2_M, IQ3_XXS. - Multimodal GGUF repos auto-bundle the vision encoder (#76) — analyzeGGUF now partitions .gguf files into LLM quants vs mmproj vision encoders. A vision_encoder SelectableItem is emitted as Recommended by default, so the recommended CLI command for a multimodal repo becomes e.g. -F q4_k_m,mm...

Read release notes

Version history

Version Updated Notes
3.0.4 Unknown What's Changed in v3.0.4 Bug-fix release tackling eight open issues reported against v3.0.3. Focused on download correctness, webui responsiveness, storage flexibility, and analyzer coverage for new quant formats. Every...
3.0.3 Unknown What's Changed in v3.0.3 Features - add GitHub Actions for releases and Docker, update to v3.0.0 (b1cf225) - add revision picker for multi-branch repos and improve web UI (183603c) - add modern web UI with analyze, cache...
3.0.1 Unknown Bug Fixes - Windows Support: Gracefully skip symlinks on Windows (downloads still work, friendly view skipped) - Installer Script: Changed -i to install command to avoid conflict with CLI's -i (interactive mode) - Exampl...
3.0.0 Unknown Release Notes - v3.0.0 Release Date: January 2026 The HuggingFace-Native Release Highlights Version 3.0.0 is a major release that brings full HuggingFace CLI compatibility. Your downloads are now stored in the standard H...
2.3.4 Unknown Updated Build script and binaries to avoid false positive virues detection for windows builds
2.3.3 Unknown 🐛 Bug Fixes Fixed: TUI/Web UI showing incorrect total file size - Issue: Total download size would start at 8GB, then jump to 35GB, then 214GB during download - Cause: plan_item events were emitted incrementally during d...
2.3.1 Unknown Release Notes - v2.3.1 Release Date: December 31, 2025 🎉 Community Features This release incorporates contributions and suggestions from the community: ✨ New Features 🚫 Exclude Patterns (--exclude, -E) Exclude specific f...
2.3.0 Unknown Release Notes - v2.3.0 Release Date: December 31, 2024 🎉 Highlights This is a major release introducing a brand new Web UI, complete project restructuring, and numerous bug fixes. The project has been reorganized into a...
2.0.0 Unknown Cleaner mental model (one download command, sensible defaults). Filesystem‑only resume—reliable and transparent; no “state” files to corrupt. JSON events and a TUI that looks great everywhere. Strong cancellation story f...
1.4.2 Unknown Forced cgo_enabled=0