Static Resumable Downloader for Hugging Face Models
ziozzang0 · reddit · 2026-07-16
The author shared hftools: a Hugging Face model downloader built as a single static Go binary, focusing on resumable downloads, checksums, and offline scenarios.
It addresses several pain points:
- Large model downloads failing midway and having to restart from scratch
- Lengthy downloads breaking easily during occasional Hub 5xx errors
- The need to verify file integrity after transferring multi-GB models to offline machines
Key features include:
- Default resumable downloads via byte offset
- --retries -1 for infinite retries during prolonged Hub outages
- File verification via Git blob SHA-1 / LFS SHA-256, with the ability to generate standard .sha256 / .sha1sum
- Conversion between "flat directory" and Hugging Face cache structures
- peek command to read safetensors / GGUF headers using a single Range request
- scan command to check for dangerous pickle imports before loading
The author notes these features stem from personal frustrations and invites feedback and comparisons with other offline verification solutions.
More from Infra
- SmolVM open-sources persistent computer infrastructure for agents that outlive chat sessions — aniketmaurya · 2026-09-11
- PyTorch Day Korea 2026 launches first offline conf, CFP closes Sept 13 — PyTorch · 2026-09-11
- Local LLM server dilemma: 4x CMP-170HX (price up 53% in 20 days) vs Mac Studio M5 Ultra — rumboll · 2026-09-11
- llama.cpp lands Flash Attention tuning for RDNA4, big prefill gains on AMD — pmttyji · 2026-09-11
- Your p99 latency benchmark may be lying: a deep dive into coordinated omission — Franc0Fernand0 · 2026-09-11
- Running MiniMax H3 on 12GB VRAM: quantization, Turbo LoRAs and attention backends compared — Possible_Mood676 · 2026-09-11