# Download speed way too slow

**URL:** </static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fdownload-speed-way-too-slow%2F169824%3E%5C
**Category:** Beginners\
**Created:** [November 2, 2025, 2:02pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fdownload-speed-way-too-slow%2F169824 "2025-11-02T14:02:09Z")\
**Posts on this page:** 12\
**Page:** 1

<div class="post-metadata">

**Author:** ![UserCase0](https://avatars.discourse-cdn.com/v4/letter/u/90ced4/32.png) [@UserCase0](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2FUserCase0)%5C
**Post date:** [November 2, 2025, 2:02pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fdownload-speed-way-too-slow%2F169824%2F1 "2025-11-02T14:02:09Z")

</div>

qwen\_image\_edit\_fp8\_e4m3fn.IRFpfsqH.safetensors.part

This takes hours. 1.9Mb/sec for 19gb is in my opinion very frustrating.

Why is this so slow, and is there a solution for this?

---

<div class="post-metadata">

**Author:** ![system](https://us1.discourse-cdn.com/hellohellohello/original/2X/d/de4155eb4aa4108ecb32a1389d7cc37ae69f88b7.png) [@system](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Fsystem)%5C
**Post date:** [November 2, 2025, 2:02pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fdownload-speed-way-too-slow%2F169824%2F2 "2025-11-02T14:02:10Z")

</div>

Unlisted on November 2, 2025, 2:02pm UTC



---

<div class="post-metadata">

**Author:** ![system](https://us1.discourse-cdn.com/hellohellohello/original/2X/d/de4155eb4aa4108ecb32a1389d7cc37ae69f88b7.png) [@system](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Fsystem)%5C
**Post date:** [November 3, 2025, 10:57am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fdownload-speed-way-too-slow%2F169824%2F3 "2025-11-03T10:57:28Z")

</div>

Listed on November 3, 2025, 10:57am UTC



---

<div class="post-metadata">

**Author:** ![John6666](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/john6666/32/27664_2.png) [@John6666](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2FJohn6666)%5C
**Post date:** [November 3, 2025, 12:44pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fdownload-speed-way-too-slow%2F169824%2F4 "2025-11-03T12:44:23Z")

</div>

Downloading via HF CLI (with XET enabled) is usually faster than using a browser, though it’s rare for browser downloads to be significantly slow…

It’s possible your ISP or network route is imposing speed restrictions.

* * *

It’s slow because you’re fetching a single 20.4 GB model file over the Hub’s Xet backend, often via one browser stream, across a congested CDN path, while your disk must assemble many chunks. That combination bottlenecks network and I/O. The specific file is indeed ~20.4 GB and stored with Xet. ([Hugging Face](https://huggingface.co/Comfy-Org/Qwen-Image_ComfyUI/blob/main/split_files/diffusion_models/qwen_image_fp8_e4m3fn.safetensors "split\_files/diffusion\_models/qwen\_image\_fp8\_e4m3fn.safetensors · Comfy-Org/Qwen-Image\_ComfyUI at main"))

# What’s happening

- **Huge artifact + Xet reconstruction.** Xet splits the file and reassembles locally. Parallel writes favor SSDs; HDDs need sequential mode or they thrash. ([Hugging Face](https://huggingface.co/docs/huggingface_hub/en/package_reference/environment_variables "Environment variables"))
- **Browser = single stream.** CLI and multi-connection tools do better than a lone HTTP stream. ([Hugging Face](https://huggingface.co/docs/huggingface_hub/en/guides/cli "Command Line Interface (CLI)"))
- **CDN region variance.** Throughput varies by location. Users report slow or unstable transfers that improve when switching region/VPN. ([Hugging Face Forums](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fdownload-instability-disconnects%2F137529 "Download instability , disconnects - Models"))
- **Observed caps.** Community reports ceilings around ~10 MB/s on some routes or setups. ([GitHub](https://github.com/Lightning-AI/litgpt/issues/1886 "Slow download from HuggingFace Hub (capped at 10.5 ..."))

# Fixes that actually speed it up

## 1) Use the HF CLI on SSD with Xet “high performance”

```bash
# Linux/macOS
pip install -U "huggingface_hub[cli]" hf_xet # docs: https://huggingface.co/docs/huggingface_hub/en/guides/cli
export HF_HOME="/fast-ssd/.cache/huggingface" # docs: https://huggingface.co/docs/huggingface_hub/en/package_reference/environment_variables
export HF_XET_HIGH_PERFORMANCE=1 # docs: https://huggingface.co/docs/huggingface_hub/en/package_reference/environment_variables
# If your download disk is an HDD, prefer sequential writes:
# export HF_XET_RECONSTRUCT_WRITE_SEQUENTIALLY=1 # same env vars doc

# Pull exactly the diffusion weight into ComfyUI
hf download Comfy-Org/Qwen-Image-Edit_ComfyUI \
  --include "split_files/diffusion_models/qwen_image_edit_fp8_e4m3fn.safetensors" \
  --local-dir "/path/to/ComfyUI/models/diffusion_models"
# CLI --local-dir reference: https://huggingface.co/docs/huggingface_hub/en/guides/cli

```

Why this helps: Xet high-performance mode increases concurrent range gets and CPU usage; cache and Xet writes land on a faster SSD; CLI avoids single-stream browser limits. ([Hugging Face](https://huggingface.co/docs/huggingface_hub/en/package_reference/environment_variables "Environment variables"))

Windows (PowerShell):

```powershell
pip install -U "huggingface_hub[cli]" hf_xet # CLI docs ↑
setx HF_HOME "D:\hf_cache" # env vars docs ↑
setx HF_XET_HIGH_PERFORMANCE 1 # env vars docs ↑
# For HDD targets only:
# setx HF_XET_RECONSTRUCT_WRITE_SEQUENTIALLY 1

hf download Comfy-Org/Qwen-Image-Edit_ComfyUI `
  --include "split_files/diffusion_models/qwen_image_edit_fp8_e4m3fn.safetensors" `
  --local-dir "C:\ComfyUI\models\diffusion_models"

```

## 2) If the CLI is still slow, try multi-connection HTTP (`aria2c`)

```bash
# Get the file’s “Copy download link” from the HF file page
# (example file page: https://huggingface.co/Comfy-Org/Qwen-Image-Edit_ComfyUI/blob/main/split_files/diffusion_models/qwen_image_edit_fp8_e4m3fn.safetensors)
aria2c -x16 -s16 -j4 -c "https://huggingface.co/Comfy-Org/Qwen-Image-Edit_ComfyUI/resolve/main/split_files/diffusion_models/qwen_image_edit_fp8_e4m3fn.safetensors?download=true"
# aria2/parallel guide: https://dev.to/susumuota/faster-and-more-reliable-hugging-face-downloads-using-aria2-and-gnu-parallel-4f2b

```

Users consistently report higher throughput with multi-connection pulls on large Hub assets. ([DEV Community](https://dev.to/susumuota/faster-and-more-reliable-hugging-face-downloads-using-aria2-and-gnu-parallel-4f2b "Faster and More Reliable Hugging Face Downloads Using ..."))

## 3) If Xet misbehaves on your route, test without it

```bash
# Temporarily force fallback from hf-xet
export HF_HUB_DISABLE_XET=1 # Windows: setx HF_HUB_DISABLE_XET 1
# Also consider bumping timeouts for flaky links
export HF_HUB_DOWNLOAD_TIMEOUT=60

```

These toggles are documented and help isolate whether the Xet path or your route is the culprit. ([Hugging Face](https://huggingface.co/docs/huggingface_hub/en/package_reference/environment_variables "Environment variables"))

## 4) Try another region path

If speed is stuck, switch network egress (VPN/off-VPN or another POP). Reports show material gains after changing region. ([Hugging Face Forums](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fdownload-instability-disconnects%2F137529 "Download instability , disconnects - Models"))

# Sanity checks for your exact case

- You are pulling the correct file and its size is expected: **20.4 GB** Xet-backed diffusion weight. If you saw a `.part` name, that’s just a partial file while downloading. Use the direct file page to confirm size before retrying. ([Hugging Face](https://huggingface.co/Comfy-Org/Qwen-Image_ComfyUI/blob/main/split_files/diffusion_models/qwen_image_fp8_e4m3fn.safetensors "split\_files/diffusion\_models/qwen\_image\_fp8\_e4m3fn.safetensors · Comfy-Org/Qwen-Image\_ComfyUI at main"))
- Use CLI or `aria2c` and verify that throughput jumps above your current ~1.9 MB/s. If not, try the VPN step and the Xet disable toggle. ([Hugging Face](https://huggingface.co/docs/huggingface_hub/en/guides/cli "Command Line Interface (CLI)"))

# References

- Qwen Image/Edit ComfyUI files. Confirms file path and 20.4 GB size for the fp8 weights. ([Hugging Face](https://huggingface.co/Comfy-Org/Qwen-Image_ComfyUI/blob/main/split_files/diffusion_models/qwen_image_fp8_e4m3fn.safetensors "split\_files/diffusion\_models/qwen\_image\_fp8\_e4m3fn.safetensors · Comfy-Org/Qwen-Image\_ComfyUI at main"))
- HF environment variables. High-performance Xet, sequential writes for HDDs, cache locations, Xet disable, and timeouts. ([Hugging Face](https://huggingface.co/docs/huggingface_hub/en/package_reference/environment_variables "Environment variables"))
- HF CLI usage, including `--local-dir`. ([Hugging Face](https://huggingface.co/docs/huggingface_hub/en/guides/cli "Command Line Interface (CLI)"))
- Reports of slow or unstable downloads and region effects. ([Hugging Face Forums](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fdownload-instability-disconnects%2F137529 "Download instability , disconnects - Models"))
- Observed caps around ~10 MB/s on some setups. ([GitHub](https://github.com/Lightning-AI/litgpt/issues/1886 "Slow download from HuggingFace Hub (capped at 10.5 ..."))
- Multi-connection download recipe that users employ to improve throughput. ([DEV Community](https://dev.to/susumuota/faster-and-more-reliable-hugging-face-downloads-using-aria2-and-gnu-parallel-4f2b "Faster and More Reliable Hugging Face Downloads Using ..."))

---

<div class="post-metadata">

**Author:** ![UserCase0](https://avatars.discourse-cdn.com/v4/letter/u/90ced4/32.png) [@UserCase0](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2FUserCase0)%5C
**Post date:** [November 19, 2025, 10:42pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fdownload-speed-way-too-slow%2F169824%2F5 "2025-11-19T22:42:31Z")

</div>

> [@John6666](#):
>
> **gle stream.** CLI and multi

Thank you John.

Your post is well documented and informative.

My internet speed at home is 500mb/sec.

I still have issue’s with slow download speeds, but not all the time.

Thanks so far !

Greetings Michel.

---

<div class="post-metadata">

**Author:** ![John6666](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/john6666/32/27664_2.png) [@John6666](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2FJohn6666)%5C
**Post date:** [November 19, 2025, 11:25pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fdownload-speed-way-too-slow%2F169824%2F6 "2025-11-19T23:25:45Z")

</div>

Well, for now, the method to set the HF CLI to the absolute fastest mode is currently as follows. Since the method for setting environment variables varies by OS, replace the `export` part according to your environment.

* * *

Below is a focused, step-by-step playbook that actually increases Hugging Face (HF) CLI download speed, with background for each step and concrete commands. The knobs and commands are current as of v1.1.x of `huggingface_hub`.

# Background: why fast ISP ≠ fast model pulls

Large files on the Hub are served via a CDN and, by default, downloaded through the HF CLI using the Rust-based `hf-xet` path. Xet reconstructs files from chunks; this favors SSD/NVMe and high concurrency, and it exposes environment variables you can tune. Route/POP (region) variability also matters, so the same 500 Mb/s line can feel “very fast” or “very slow” depending on path and local I/O. ([Hugging Face](https://huggingface.co/docs/huggingface_hub/main/en/package_reference/environment_variables "Environment variables"))

* * *

# 1) Put the cache on a fast SSD and update the tools

**Why:** Eliminates a common disk bottleneck and ensures you have recent CLI/Xet behavior. `HF_HOME` controls where the Hub cache lives. ([Hugging Face](https://huggingface.co/docs/huggingface_hub/main/en/package_reference/environment_variables "Environment variables"))

```bash
# References:
# - HF env vars (HF_HOME, HF_HUB_CACHE, timeouts, Xet knobs): https://huggingface.co/docs/huggingface_hub/en/package_reference/environment_variables
# - CLI install/usage: https://huggingface.co/docs/huggingface_hub/en/guides/cli
python -m pip install -U "huggingface_hub" hf_xet
export HF_HOME="/fast-ssd/.cache/huggingface" # move cache to SSD/NVMe
export HF_HUB_DOWNLOAD_TIMEOUT=60 # reduce spurious timeouts on slow routes

```

`HF_HOME` sets the base cache dir; `HF_HUB_DOWNLOAD_TIMEOUT` raises the per-request timeout from the 10 s default to something more forgiving. ([Hugging Face](https://huggingface.co/docs/huggingface_hub/main/en/package_reference/environment_variables "Environment variables"))

* * *

# 2) Turn on Xet “high performance” and set explicit concurrency

**Why:** The Xet backend exposes controls that directly impact throughput. High-perf mode tries to saturate network/CPU; the per-file concurrency (`HF_XET_NUM_CONCURRENT_RANGE_GETS`) increases parallel chunk reads. Defaults are conservative (16). ([Hugging Face](https://huggingface.co/docs/huggingface_hub/main/en/package_reference/environment_variables "Environment variables"))

```bash
# References (HF_XET_HIGH_PERFORMANCE, HF_XET_NUM_CONCURRENT_RANGE_GETS):
# https://huggingface.co/docs/huggingface_hub/en/package_reference/environment_variables
export HF_XET_HIGH_PERFORMANCE=1
export HF_XET_NUM_CONCURRENT_RANGE_GETS=24 # try 24–32 on fast SSD/CPU; 8–16 on weaker machines

```

These variables are officially documented; raising concurrency helps if you have headroom. ([Hugging Face](https://huggingface.co/docs/huggingface_hub/main/en/package_reference/environment_variables "Environment variables"))

* * *

# 3) If the destination is an HDD, force sequential writes

**Why:** Parallel random writes that help SSDs can thrash spinning disks. The Xet toggle below switches to sequential writes and is specifically recommended for HDDs. ([Hugging Face](https://huggingface.co/docs/huggingface_hub/main/en/package_reference/environment_variables "Environment variables"))

```bash
# Reference: HF_XET_RECONSTRUCT_WRITE_SEQUENTIALLY (HDD-only)
# https://huggingface.co/docs/huggingface_hub/en/package_reference/environment_variables
export HF_XET_RECONSTRUCT_WRITE_SEQUENTIALLY=1

```

* * *

# 4) Download only what you need; use `--include` and `--local-dir`

**Why:** Avoids pulling entire repos and keeps writes local to your target folder. The CLI implements patterns and a direct “download-to-folder” mode. ([Hugging Face](https://huggingface.co/docs/huggingface_hub/en/guides/cli "Command Line Interface (CLI)"))

```bash
# References:
# - CLI download patterns & --local-dir: https://huggingface.co/docs/huggingface_hub/en/guides/cli
# - Download guide (hf_hub_download/snapshot_download): https://huggingface.co/docs/huggingface_hub/en/guides/download
hf download Comfy-Org/Qwen-Image-Edit_ComfyUI \
  --include "split_files/diffusion_models/qwen_image_edit_fp8_e4m3fn.safetensors" \
  --local-dir "/path/to/ComfyUI/models/diffusion_models"

```

* * *

# 5) If speeds are spiky under the CLI, A/B test the transfer path

**Why:** Isolate whether Xet routing/behavior is the issue on your machine/route. You can explicitly disable Xet; set this **before** importing or invoking HF code so it’s honored. ([Hugging Face](https://huggingface.co/docs/huggingface_hub/main/en/package_reference/environment_variables "Environment variables"))

```bash
# References:
# - HF_HUB_DISABLE_XET (must be set prior to import/CLI): https://huggingface.co/docs/huggingface_hub/en/package_reference/environment_variables
# - Note about setting it before use: https://github.com/huggingface/huggingface_hub/issues/3266
export HF_HUB_DISABLE_XET=1
hf download <repo> --include "<file>" --local-dir "<dir>"

```

* * *

# 6) Use a multi-connection HTTP downloader for one-off big pulls

**Why:** When a single TCP stream underperforms on your current path, multiple HTTP range connections often achieve higher steady throughput. `aria2c` is a simple, robust option. ([DEV Community](https://dev.to/susumuota/faster-and-more-reliable-hugging-face-downloads-using-aria2-and-gnu-parallel-4f2b "Faster and More Reliable Hugging Face Downloads Using ..."))

```bash
# Reference/recipe: https://dev.to/susumuota/faster-and-more-reliable-hugging-face-downloads-using-aria2-and-gnu-parallel-4f2b
# 1) Get the “Copy download link” (?download=true) from the file’s page on the Hub
aria2c -x16 -s16 -j4 -c \
 "https://huggingface.co/<repo>/resolve/main/<path/to/large-file>?download=true"
# -x connections per server; -s splits per file; -j parallel jobs; -c resume

```

* * *

# 7) Change network egress (region) when speeds are unusually low

**Why:** The Hub’s CDN and peering vary by POP and time. Toggling VPN on/off or switching to a nearby region often changes anemic 1–2 MB/s to tens of MB/s. Forum reports and timeout traces on `cdn-lfs*.hf.co` illustrate this effect. ([Hugging Face Forums](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fdownloads-intermittently-fail-403-low-bandwidth%2F140198 "Downloads intermittently fail, 403, low bandwidth"))

* * *

# 8) Know what’s deprecated (so you don’t chase the wrong lever)

**Why:** Many older posts recommend `hf_transfer`. In current `huggingface_hub` v1.x, **`hf_transfer` is removed and `HF_HUB_ENABLE_HF_TRANSFER` is ignored**. Use the Xet knobs (`HF_XET_HIGH_PERFORMANCE`, etc.) instead. ([Hugging Face](https://huggingface.co/docs/huggingface_hub/en/concepts/migration "Migrating to huggingface\_hub v1.0"))

* * *

## 60-second diagnosis (copy/paste)

This quickly tells you where the bottleneck is on your system/route.

```bash
# 0) Prep
python -m pip install -U "huggingface_hub" hf_xet # https://huggingface.co/docs/huggingface_hub/en/guides/cli
export HF_HOME="/fast-ssd/.cache/huggingface" # https://huggingface.co/docs/huggingface_hub/en/package_reference/environment_variables

# 1) CLI with tuned Xet (expect best performance on SSD)
export HF_XET_HIGH_PERFORMANCE=1 # https://huggingface.co/docs/huggingface_hub/en/package_reference/environment_variables
export HF_XET_NUM_CONCURRENT_RANGE_GETS=24
hf download <repo> --include "<bigfile>" --local-dir "<dir>"

# 2) Same CLI, but Xet off (if #1 was spiky)
export HF_HUB_DISABLE_XET=1 # set BEFORE running HF code
hf download <repo> --include "<bigfile>" --local-dir "<dir>"

# 3) Multi-connection HTTP (expect steady speeds if single-stream path is weak)
aria2c -x16 -s16 -j4 -c "<direct ?download=true URL>" # https://dev.to/susumuota/faster-and-more-reliable-hugging-face-downloads-using-aria2-and-gnu-parallel-4f2b

```

**Interpretation:**  
• #1 fast and stable → keep Xet high-perf + SSD.  
• #1 spiky but #2 steady → Xet/route interaction on this box; run with Xet disabled here. ([GitHub](https://github.com/huggingface/huggingface_hub/issues/3266 "HF\_HUB\_DISABLE\_XET not disabling XET-based ..."))  
• Both slow but `aria2c` fast → single-stream/POP issue; keep CLI for convenience but prefer multi-connection pulls for huge files. ([DEV Community](https://dev.to/susumuota/faster-and-more-reliable-hugging-face-downloads-using-aria2-and-gnu-parallel-4f2b "Faster and More Reliable Hugging Face Downloads Using ..."))  
• All slow → switch region (VPN/off-VPN) and retry. ([Hugging Face Forums](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fdownloads-intermittently-fail-403-low-bandwidth%2F140198 "Downloads intermittently fail, 403, low bandwidth"))

* * *

## Common pitfalls (quick checks)

- **Downloading whole repos by accident.** Use `--include` to pull only the large file(s) you need. ([Hugging Face](https://huggingface.co/docs/huggingface_hub/en/guides/cli "Command Line Interface (CLI)"))
- **Tiny default timeouts.** Raise `HF_HUB_DOWNLOAD_TIMEOUT` to reduce intermittent read timeouts on `cdn-lfs*.hf.co`. ([Hugging Face](https://huggingface.co/docs/huggingface_hub/en/guides/cli "Command Line Interface (CLI)"))
- **HDD as destination.** If you must, add `HF_XET_RECONSTRUCT_WRITE_SEQUENTIALLY=1`. ([Hugging Face](https://huggingface.co/docs/huggingface_hub/main/en/package_reference/environment_variables "Environment variables"))
- **Expecting `hf_transfer` to help.** It’s gone in v1.x; use Xet high-perf instead. ([Hugging Face](https://huggingface.co/docs/huggingface_hub/en/concepts/migration "Migrating to huggingface\_hub v1.0"))

* * *

## Short, curated references

**Official docs**

- Environment variables (HF\_HOME/HF\_HUB\_CACHE, timeouts, Xet knobs, disable Xet). Useful to confirm every variable above. ([Hugging Face](https://huggingface.co/docs/huggingface_hub/main/en/package_reference/environment_variables "Environment variables"))
- CLI guide (`hf download`, `--include`, `--local-dir`, timeouts). Step-by-step usage examples. ([Hugging Face](https://huggingface.co/docs/huggingface_hub/en/guides/cli "Command Line Interface (CLI)"))
- Migration note: `hf_transfer` removed; `HF_HUB_ENABLE_HF_TRANSFER` ignored; use `HF_XET_HIGH_PERFORMANCE`. ([Hugging Face](https://huggingface.co/docs/huggingface_hub/en/concepts/migration "Migrating to huggingface\_hub v1.0"))

**GitHub issues (real-world behavior)**

- “HF\_HUB\_DISABLE\_XET not disabling unless set before import” — clarifies when the toggle takes effect. ([GitHub](https://github.com/huggingface/huggingface_hub/issues/3266 "HF\_HUB\_DISABLE\_XET not disabling XET-based ..."))
- “Large downloads stuck/slow with VPN” — intermittent slow/0% cases on big files with CLI. ([GitHub](https://github.com/huggingface/huggingface_hub/issues/2691 "downloading large files got stuck at 0% forever when using ..."))

**Community threads/recipes**

- Intermittent slow speeds, disconnects, and region changes as fixes. Background on route/POP sensitivity. ([Hugging Face Forums](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fdownloads-intermittently-fail-403-low-bandwidth%2F140198 "Downloads intermittently fail, 403, low bandwidth"))
- Practical `aria2c` recipe for multi-connection, resumable downloads. ([DEV Community](https://dev.to/susumuota/faster-and-more-reliable-hugging-face-downloads-using-aria2-and-gnu-parallel-4f2b "Faster and More Reliable Hugging Face Downloads Using ..."))

---

<div class="post-metadata">

**Author:** ![ibraheemzia](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/ibraheemzia/32/55862_2.png) [@ibraheemzia](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Fibraheemzia)%5C
**Post date:** [November 21, 2025, 3:21am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fdownload-speed-way-too-slow%2F169824%2F7 "2025-11-21T03:21:16Z")

</div>

It sounds like a lot of users are running into the same slowdown, so the issue is likely on HF’s end rather than your setup. Sometimes the CDN gets overloaded, especially during peak hours or large model releases. You can try switching to a different network or using a download manager, but honestly, the speed usually returns to normal after a while. Hopefully Hugging Face resolves the bottleneck soon.

---

<div class="post-metadata">

**Author:** ![mootari](https://avatars.discourse-cdn.com/v4/letter/m/34f0e0/32.png) [@mootari](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Fmootari)%5C
**Post date:** [December 21, 2025, 12:26pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fdownload-speed-way-too-slow%2F169824%2F10 "2025-12-21T12:26:28Z")

</div>

`HF_HUB_DISABLE_XET=1` did the trick for me. I had installed huggingface-cli via Homebrew, and without the flag the transfer speeds for files stored in xet would cap at around 290 kb/s. I went back and forth to verify that the difference was reproducible.

Edit: I also had to delete the cache since the prior attempts resulted in a corrupt file.

---

<div class="post-metadata">

**Author:** ![jwaters8978](https://avatars.discourse-cdn.com/v4/letter/j/df705f/32.png) [@jwaters8978](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Fjwaters8978)%5C
**Post date:** [April 9, 2026, 5:49am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fdownload-speed-way-too-slow%2F169824%2F12 "2026-04-09T05:49:52Z")

</div>

The problem might be that you aren’t using parquet format. I just downloaded a dataset with with 10k items (3GB) and it took 2 hours and was averaging 30mbps in 1 gbit internet.

Downloading in parquet fixed it and I got full download throughput. Most datasets have an auto parquet converter, so you can use that ref to get the fast downloads.

Works much faster for access and uploading too! parquet rocks

---

<div class="post-metadata">

**Author:** ![khoinguyenbk](https://avatars.discourse-cdn.com/v4/letter/k/c6cbf5/32.png) [@khoinguyenbk](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Fkhoinguyenbk)%5C
**Post date:** [April 13, 2026, 4:01pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fdownload-speed-way-too-slow%2F169824%2F13 "2026-04-13T16:01:59Z")

</div>

So basically, HF acquired XetHub just to have a new performance regression? 😆

---

<div class="post-metadata">

**Author:** ![kvcachemiss](https://avatars.discourse-cdn.com/v4/letter/k/e79b87/32.png) [@kvcachemiss](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Fkvcachemiss)%5C
**Post date:** [May 14, 2026, 7:02pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fdownload-speed-way-too-slow%2F169824%2F14 "2026-05-14T19:02:26Z")

</div>

This resolved the slowdown issues for me as well.

---

<div class="post-metadata">

**Author:** ![mindph](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/mindph/32/51902_2.png) [@mindph](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Fmindph)%5C
**Post date:** [October 8, 2026, 8:17pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fdownload-speed-way-too-slow%2F169824%2F15 "2026-10-08T20:17:58Z")

</div>

The download speed is unbearable I get down to below 1mb / s when using git checkout to download a branch. No I do not want to install Python modules to download a model, I have Git. Git works everywhere. HF is just slow.
