# Cas-bridge.xethub.hf.co broke **URL:** \ **Category:** 🤗Hub\ **Created:** [June 9, 2025, 5:30pm UTC](https://discuss.huggingface.co/t/cas-bridge-xethub-hf-co-broke/158626 "2025-06-09T17:30:44Z")\ **Posts on this page:** 15\ **Page:** 1

**Author:** ![Kostya165](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/kostya165/32/49136_2.png) [@Kostya165](https://discuss.huggingface.co/u/Kostya165)\ **Post date:** [June 9, 2025, 5:30pm UTC](https://discuss.huggingface.co/t/cas-bridge-xethub-hf-co-broke/158626/1 "2025-06-09T17:30:44Z")

Hello everyone, literally 5-6 hours ago I started encountering questionable errors. Models and datasets refuse to download, even those that I have published.Disabling XET did not help. Reinstalling libraries, sudo apt upgrade/update, reinstalling certificates too.Reinstalling libraries, sudo apt upgrade/update, reinstalling certificates too. I found that the problem is [https://cas-bridge.xethub.hf.co](https://cas-bridge.xethub.hf.co) - it does not respond to the curl request. Has anyone ever encountered this? ---

**Author:** ![John6666](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/john6666/32/27664_2.png) [@John6666](https://discuss.huggingface.co/u/John6666)\ **Post date:** [June 10, 2025, 5:12am UTC](https://discuss.huggingface.co/t/cas-bridge-xethub-hf-co-broke/158626/2 "2025-06-10T05:12:33Z")

Domain (false) blacklisting issue…? > [@How to get a list of all Huggingface download redirections to whitelist?](https://discuss.huggingface.co/t/how-to-get-a-list-of-all-huggingface-download-redirections-to-whitelist/30486): > > I work inside a secure corporate VPN network, so I’m unable to download Huggingface models using from\_pretrained commands. However, I can request the security team to whitelist certain URLs needed for my use-case. The security team has already whitelisted the ‘[huggingface.co](http://huggingface.co)’ and ‘[cdn-lfs.huggingface.co](http://cdn-lfs.huggingface.co)’ URLs. I can now download the files from repo but the loading functions from\_pretrained still don’t work. I think it’s getting blocked while redirecting the requests internally. So, is there a … ---

**Author:** ![Kostya165](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/kostya165/32/49136_2.png) [@Kostya165](https://discuss.huggingface.co/u/Kostya165)\ **Post date:** [June 10, 2025, 6:35am UTC](https://discuss.huggingface.co/t/cas-bridge-xethub-hf-co-broke/158626/3 "2025-06-10T06:35:43Z")

I dont think so, i dont work in secure corporate VPN network:) ---

**Author:** ![jsulz](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/jsulz/32/28279_2.png) [@jsulz](https://discuss.huggingface.co/u/jsulz)\ **Post date:** [June 10, 2025, 10:32pm UTC](https://discuss.huggingface.co/t/cas-bridge-xethub-hf-co-broke/158626/4 "2025-06-10T22:32:29Z")

Hi @Kostya165 - Xet team member here; so sorry to hear you’re running into problems! Could you provide more information about what you’re experiencing? Specifically: - Some example repos where you’re experiencing this. - What APIs you’re using to download with (e.g., `huggingface_hub`’s `snapshot_download` or `transformers`’s `from_pretrained`, etc); or, if you’re using the web browser/curl, some information about that (e.g., what web browser) - When you were using `hf-xet`, what version of the package where you using? A full output of `pip freeze` would be helpful. - What platform/OS you’re using - Any error statements or logs ---

**Author:** ![hadjuse](https://avatars.discourse-cdn.com/v4/letter/h/41988e/32.png) [@hadjuse](https://discuss.huggingface.co/u/hadjuse)\ **Post date:** [June 17, 2025, 1:21pm UTC](https://discuss.huggingface.co/t/cas-bridge-xethub-hf-co-broke/158626/5 "2025-06-17T13:21:01Z")

Hello I use jetbrains my python version is 3.11 It doesnt work with NVILA model: > **[Efficient-Large-Model/NVILA-8B · Hugging Face](https://huggingface.co/Efficient-Large-Model/NVILA-8B)** > > We’re on a journey to advance and democratize artificial intelligence through open source and open science. I would like to share my code here and I’m currently in entreprise VPN: ```python import os os.environ['REQUESTS_CA_BUNDLE'] = r'mycert.pem' os.environ['CURL_CA_BUNDLE'] = os.environ['REQUESTS_CA_BUNDLE'] #os.environ['HF_HUB_DISABLE_SSL_VERIFY'] = '1' from huggingface_hub import snapshot_download from transformers import AutoTokenizer, AutoModelForCausalLM # --- CONFIGURATION --- REPO_ID = "Efficient-Large-Model/NVILA-8B-Video" LOCAL_DIR = os.path.abspath(r'model_nvila_8B') if not os.path.exists(LOCAL_DIR): snapshot_download( repo_id=REPO_ID, repo_type="model", local_dir=LOCAL_DIR, ) else: print("Model already downloaded") ``` please find out log: ```json {"timestamp":"2025-06-17T13:18:14.390107Z","level":"WARN","fields":{"message":"Reqwest(reqwest::Error { kind: Request, url: \"https://cas-server.xethub.hf.co/reconstruction/688bce9d9130b743bf92cda612964528312d5d3c97a990e7c3e4a9db97428c6f\", source: hyper_util::client::legacy::Error(Connect, Custom { kind: InvalidData, error: InvalidCertificate(UnknownIssuer) }) }). Retrying..."},"filename":"D:\\a\\xet-core\\xet-core\\cas_client\\src\\http_client.rs","line_number":242} ``` Stacktrace python ```auto RuntimeError: Data processing error: CAS service error : ReqwestMiddleware Error: Request failed after 5 retries ``` ---

**Author:** ![jsulz](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/jsulz/32/28279_2.png) [@jsulz](https://discuss.huggingface.co/u/jsulz)\ **Post date:** [June 18, 2025, 9:31pm UTC](https://discuss.huggingface.co/t/cas-bridge-xethub-hf-co-broke/158626/6 "2025-06-18T21:31:22Z")

HI @hadjuse I believe this is likely a similar issue to [Xet hub InvalidCertificate(UnknownIssuer) · Issue #389 · huggingface/xet-core · GitHub](https://github.com/huggingface/xet-core/issues/389) We’ll investigate there and follow up here, but feel free to add your report to that issue as well so you can follow along and get updates. ---

**Author:** ![hoytak](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/hoytak/32/49686_2.png) [@hoytak](https://discuss.huggingface.co/u/hoytak)\ **Post date:** [June 18, 2025, 11:29pm UTC](https://discuss.huggingface.co/t/cas-bridge-xethub-hf-co-broke/158626/7 "2025-06-18T23:29:00Z")

Hello! Thanks for the report. We introduced some changes to the cert loading to fix some other reported issues, but it seems that it wasn’t fully robust. I added a few changes in to try to ensure these cases are covered too. Could you all see if `pip install hf_xet==v1.1.5-rc1` fixes your issues @Kostya165@John6666@hadjuse ? ---

**Author:** ![John6666](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/john6666/32/27664_2.png) [@John6666](https://discuss.huggingface.co/u/John6666)\ **Post date:** [June 19, 2025, 12:11am UTC](https://discuss.huggingface.co/t/cas-bridge-xethub-hf-co-broke/158626/8 "2025-06-19T00:11:10Z")

Seems working well for me. (Windows (not WSL2), Python 3.9) ```auto pip install -U huggingface_hub[hf_xet] ``` ```py import os from huggingface_hub import snapshot_download from transformers import AutoTokenizer, AutoModelForCausalLM # --- CONFIGURATION --- REPO_ID = "Efficient-Large-Model/NVILA-8B-Video" LOCAL_DIR = os.path.abspath(r'model_nvila_8B') if not os.path.exists(LOCAL_DIR): snapshot_download( repo_id=REPO_ID, repo_type="model", local_dir=LOCAL_DIR, ) else: print("Model already downloaded") ``` ```auto README.md: 100%|████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 4.29k/4.29k [00:00<00:00, 2.14MB/s] added_tokens.json: 100%|████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 194/194 [00:00<00:00, 64.6kB/s] generation_config.json: 100%|████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 243/243 [00:00<00:00, 122kB/s] config.json: 100%|████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 226k/226k [00:00<00:00, 8.70MB/s] merges.txt: 100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 1.67M/1.67M [00:00<00:00, 21.1MB/s] config.json: 100%|████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 186k/186k [00:00<00:00, 1.19MB/s] .gitattributes: 100%|████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 1.52k/1.52k [00:00<00:00, 759kB/s] model.safetensors.index.json: 100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 27.8k/27.8k [00:00<00:00, 9.25MB/s] special_tokens_map.json: 100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 555/555 [00:00<00:00, 277kB/s] tokenizer_config.json: 100%|████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 2.30k/2.30k [00:00<00:00, 1.15MB/s] config.json: 100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 313/313 [00:00<00:00, 156kB/s] trainer_state.json: 100%|██████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 128k/128k [00:00<00:00, 860kB/s] config.json: 100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 650/650 [00:00<00:00, 217kB/s] vocab.json: 100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 3.38M/3.38M [00:00<00:00, 4.32MB/s] preprocessor_config.json: 100%|██████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 394/394 [00:00<00:00, 131kB/s] mm_projector/model.safetensors: 100%|████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 58.8M/58.8M [01:14<00:00, 789kB/s] llm/model-00001-of-00004.safetensors: 0%| | 3.39M/4.87G [01:11<29:12:40, 46.3kB/s] llm/model-00003-of-00004.safetensors: 0%| | 2.27M/4.33G [00:38<9:42:44, 124kB/s] llm/model-00004-of-00004.safetensors: 7%|█████████▋ | 81.2M/1.09G [01:10<13:31, 1.24MB/s] llm/model-00004-of-00004.safetensors: 14%|█████████████████▋ | 148M/1.09G [01:14<06:11, 2.53MB/s] vision_tower/model.safetensors: 4%|████▉ | 30.4M/827M [01:08<1:09:24, 191kB/s] mm_projector/model.safetensors: 100%|████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 58.8M/58.8M [01:14<00:00, 789kB/s] ``` ---

**Author:** ![hoytak](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/hoytak/32/49686_2.png) [@hoytak](https://discuss.huggingface.co/u/hoytak)\ **Post date:** [June 20, 2025, 9:56pm UTC](https://discuss.huggingface.co/t/cas-bridge-xethub-hf-co-broke/158626/9 "2025-06-20T21:56:55Z")

Hello everyone – we just released a new version of hf\_xet to address this issue. Please upgrade to the newest release (`pip install --upgrade hf_xet==v1.1.5`) if you encounter any of these issues. ---

**Author:** ![dakerholdings](https://avatars.discourse-cdn.com/v4/letter/d/45deac/32.png) [@dakerholdings](https://discuss.huggingface.co/u/dakerholdings)\ **Post date:** [September 5, 2025, 6:01pm UTC](https://discuss.huggingface.co/t/cas-bridge-xethub-hf-co-broke/158626/10 "2025-09-05T18:01:09Z")

I was just getting a similar set of these “{“timestamp”:…“level”:“WARN”,“fields”:{“message”:“Reqwest(reqwest::Error { kind: Request, url: “[https://transfer.xethub.hf.co/xorbs/default/38f44618480f1bf7552fb8c9cda5dfa5b6fb59378db0f2513bc066beb34e2075?X-Xet-Signed-Range=bytes%3D0-67053234&Expires=1757097156&Policy=…”](https://transfer.xethub.hf.co/xorbs/default/38f44618480f1bf7552fb8c9cda5dfa5b6fb59378db0f2513bc066beb34e2075?X-Xet-Signed-Range=bytes%3D0-67053234&Expires=1757097156&Policy=%E2%80%A6%E2%80%9D), source: hyper\_util::client::legacy::Error(Connect, Error { code: -9805, message: “connection closed gracefully” }) }). Retrying…“},“filename”:”/Users/runner/work/xet-core/xet-core/cas\_client/src/http\_client.rs”,“line\_number”:200}” error messages; after reading the above thread, I tried upgrading hf\_xet (which on my system had apparently remained at 1.1.3!? why no upgrade requirements, I wonder?: ds@Ds-MacBook-Air ~ % pip install --upgrade hf\_xet Requirement already satisfied: hf\_xet in ./.pyenv/versions/3.10.4/lib/python3.10/site-packages (1.1.3) Collecting hf\_xet Downloading hf\_xet-1.1.9-cp37-abi3-macosx\_11\_0\_arm64.whl.metadata (4.7 kB) … Now the latest model I was attempting to download has thus far been downloading without errors (although I had interrupted it with gigs of progress having been made already, and that progress was apparently not resumed in situ, instead starting from very near the beginning). I think that any interruptions should be able to retain a reasonably maximal amount of progress [with the previous version of hf\_xet, thus far interruptions have appeared to lose significant minority percentages of progress already made!?]. ---

**Author:** ![dakerholdings](https://avatars.discourse-cdn.com/v4/letter/d/45deac/32.png) [@dakerholdings](https://discuss.huggingface.co/u/dakerholdings)\ **Post date:** [September 6, 2025, 1:04am UTC](https://discuss.huggingface.co/t/cas-bridge-xethub-hf-co-broke/158626/11 "2025-09-06T01:04:12Z")

Hmm; in continuing with that download, now I’ve been receiving a different kind of xet error message: 2025-09-06T00:55:15.588106Z ERROR Fatal Error: “s3::get\_range” api call failed (request id , retry 2): HTTP status client error (403 Forbidden) for url ([https://transfer.xethub.hf.co/xorbs/default/…?X-Xet-Signed-Range=bytes%3D0-67070887&Expires=1757117149&Policy=…&Signature=…](https://transfer.xethub.hf.co/xorbs/default/%E2%80%A6?X-Xet-Signed-Range=bytes%3D0-67070887&Expires=1757117149&Policy=%E2%80%A6&Signature=%E2%80%A6)) And, network traffic also went → to 0, near completion; I wonder if there may be any key or signal or other way currently available to prod it into re-starting the network traffic without explicit interruption? ---

**Author:** ![John6666](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/john6666/32/27664_2.png) [@John6666](https://discuss.huggingface.co/u/John6666)\ **Post date:** [September 6, 2025, 1:30am UTC](https://discuss.huggingface.co/t/cas-bridge-xethub-hf-co-broke/158626/12 "2025-09-06T01:30:46Z")

Hmm… [Specifying the cache size](https://github.com/huggingface/xet-core/issues/311#issuecomment-2896529948) may alleviate the symptoms in some cases, but if the issue persists, it might be quicker to raise an issue… ---

**Author:** ![dakerholdings](https://avatars.discourse-cdn.com/v4/letter/d/45deac/32.png) [@dakerholdings](https://discuss.huggingface.co/u/dakerholdings)\ **Post date:** [September 7, 2025, 2:40am UTC](https://discuss.huggingface.co/t/cas-bridge-xethub-hf-co-broke/158626/13 "2025-09-07T02:40:19Z")

I’m not sure that it’s the same issue (or maybe there’s more than one); if I instead “export HF\_HUB\_ENABLE\_HF\_TRANSFER=1”, then only one file gets downloaded at a time (and no errors shown during xfer), but still eventually network traffic drops to zero (seeming to happen at over 4 gigs into xfer), with the interrupt stack trace showing: model-00004-of-00005.safetensors: 95%|████████ | 4.99G/5.26G [1:03:32\<02:49, 1.58MB/s]^CCancellation requested; stopping current tasks. Traceback (most recent call last): File “/Users/ds/.pyenv/versions/3.10.4/lib/python3.10/site-packages/huggingface\_hub/file\_download.py”, line 629, in xet\_get download\_files( KeyboardInterrupt During handling of the above exception, another exception occurred: Traceback (most recent call last): File “/Users/ds/.pyenv/versions/3.10.4/bin/mlx\_lm.convert”, line 7, in sys.exit(main()) File “/Users/ds/.pyenv/versions/3.10.4/lib/python3.10/site-packages/mlx\_lm/convert.py”, line 224, in main convert(\*\*vars(args)) File “/Users/ds/.pyenv/versions/3.10.4/lib/python3.10/site-packages/mlx\_lm/convert.py”, line 110, in convert model\_path, hf\_path = get\_model\_path(hf\_path, revision=revision) File “/Users/ds/.pyenv/versions/3.10.4/lib/python3.10/site-packages/mlx\_lm/utils.py”, line 102, in get\_model\_path snapshot\_download( File “/Users/ds/.pyenv/versions/3.10.4/lib/python3.10/site-packages/huggingface\_hub/utils/\_validators.py”, line 114, in \_inner\_fn return fn(\*args, \*\*kwargs) File “/Users/ds/.pyenv/versions/3.10.4/lib/python3.10/site-packages/huggingface\_hub/\_snapshot\_download.py”, line 330, in snapshot\_download \_inner\_hf\_hub\_download(file) File “/Users/ds/.pyenv/versions/3.10.4/lib/python3.10/site-packages/huggingface\_hub/\_snapshot\_download.py”, line 306, in \_inner\_hf\_hub\_download return hf\_hub\_download( File “/Users/ds/.pyenv/versions/3.10.4/lib/python3.10/site-packages/huggingface\_hub/utils/\_validators.py”, line 114, in \_inner\_fn return fn(\*args, \*\*kwargs) File “/Users/ds/.pyenv/versions/3.10.4/lib/python3.10/site-packages/huggingface\_hub/file\_download.py”, line 1010, in hf\_hub\_download return \_hf\_hub\_download\_to\_cache\_dir( File “/Users/ds/.pyenv/versions/3.10.4/lib/python3.10/site-packages/huggingface\_hub/file\_download.py”, line 1171, in \_hf\_hub\_download\_to\_cache\_dir \_download\_to\_tmp\_and\_move( File “/Users/ds/.pyenv/versions/3.10.4/lib/python3.10/site-packages/huggingface\_hub/file\_download.py”, line 1723, in \_download\_to\_tmp\_and\_move xet\_get( File “/Users/ds/.pyenv/versions/3.10.4/lib/python3.10/site-packages/huggingface\_hub/file\_download.py”, line 629, in xet\_get download\_files( KeyboardInterrupt model-00004-of-00005.safetensors: 95%|██████ | 4.99G/5.26G [1:29:51\<04:49, 926kB/s] ---

**Author:** ![jsulz](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/jsulz/32/28279_2.png) [@jsulz](https://discuss.huggingface.co/u/jsulz)\ **Post date:** [September 9, 2025, 4:26pm UTC](https://discuss.huggingface.co/t/cas-bridge-xethub-hf-co-broke/158626/14 "2025-09-09T16:26:34Z")

Thanks for the stack trace @dakerholdings! Getting a `403` for a URL [like the one listed here](https://discuss.huggingface.co/t/cas-bridge-xethub-hf-co-broke/158626/11) indicates that the URL for the content has expired. Could you tell me a bit more about your download/environment? - What repository you are trying to download - Your network speed - The output from `hf env` (that provides details about packages and your current platform) - A minimal reproducible script that shows what APIs/commands you are using when you encounter this problem Thank you! ---

**Author:** ![n30tri8](https://avatars.discourse-cdn.com/v4/letter/n/3be4f8/32.png) [@n30tri8](https://discuss.huggingface.co/u/n30tri8)\ **Post date:** [November 1, 2025, 12:20pm UTC](https://discuss.huggingface.co/t/cas-bridge-xethub-hf-co-broke/158626/15 "2025-11-01T12:20:27Z")

I have the same issue downloading a llama model on vertex ai environment. - repo I want to download: meta-llama/Llama-3.1-8B-Instruct - run environment: vertex ai, custom training job, custom docker details: - running the app in a container on my local pc, I can download the model, although “Killled” after since low memory. - the command which the app gets stuck in vertex ai: model = AutoModelForCausalLM.from\_pretrained( ‘meta-llama/Llama-3.1-8B-Instruct’, dtype=torch.float16, cache\_dir=HF\_CACHE\_DIR, low\_cpu\_mem\_usage=True ) error log: 31mFatal Error: \“s3::get\_range\” api call failed (request id , retry 2): HTTP status client error (403 Forbidden) for url ([https://transfer.xethub.hf.co/xorbs/default/9fa9…](https://transfer.xethub.hf.co/xorbs/default/9fa9%E2%80%A6) edit: I cache the model on google cloud storage, as specified in the command; and I checked the size of the folder of the model: 14.97 GiB, which is similar to the full model on my os. Still, my app is stuck with the mentioned error.