I want to download 3.1-70B-Instruct using the python llama stack client. When downloading the 8B-Instruct Model it works fine. Other tested models also work fine. 70B-Instruct fails on downloading consolidated.01.pth, consolidated00.pth` runs through. Also see the scrollback.
Environment is WSL, 24.04 LTS, 3090 GPU
Steps to reproduce
python3 -m venv .venv && source .venv/bin/activate, llama download --source meta --model-id Llama3.1-70B-Instruct, paste url and wait
Expected behaviour
successful download
Actual behaviour
download fails on consolidated.01.pth with Error 405
Scrollback
llama download --source meta --model-id Llama3.1-70B-Instruct
Please provide the signed URL you received via email (e.g., https://llama3-1.llamameta.net/*?Policy...):
<... redacted ...>
Downloading `checklist.chk`...
Already downloaded `/home/nano/.llama/checkpoints/Llama3.1-70B-Instruct/checklist.chk`, skipping...
Downloading `tokenizer.model`...Already downloaded `/home/nano/.llama/checkpoints/Llama3.1-70B-Instruct/tokenizer.model`, skipping...
Downloading `params.json`...
Already downloaded `/home/nano/.llama/checkpoints/Llama3.1-70B-Instruct/params.json`, skipping...
Downloading `consolidated.00.pth`...
Already downloaded `/home/nano/.llama/checkpoints/Llama3.1-70B-Instruct/consolidated.00.pth`, skipping...
Downloading `consolidated.01.pth`...
Traceback (most recent call last):
File "/mnt/d/repos/llama/llama3/.venv/bin/llama", line 8, in <module>
sys.exit(main())
^^^^^^
File "/mnt/d/repos/llama/llama3/.venv/lib/python3.12/site-packages/llama_stack/cli/llama.py", line 44, in main
parser.run(args)
File "/mnt/d/repos/llama/llama3/.venv/lib/python3.12/site-packages/llama_stack/cli/llama.py", line 38, in run
args.func(args)
File "/mnt/d/repos/llama/llama3/.venv/lib/python3.12/site-packages/llama_stack/cli/download.py", line 174, in run_download_cmd
_meta_download(model, meta_url)
File "/mnt/d/repos/llama/llama3/.venv/lib/python3.12/site-packages/llama_stack/cli/download.py", line 143, in _meta_download
asyncio.run(downloader.download())
File "/usr/lib/python3.12/asyncio/runners.py", line 194, in run
return runner.run(main)
^^^^^^^^^^^^^^^^
File "/usr/lib/python3.12/asyncio/runners.py", line 118, in run
return self._loop.run_until_complete(task)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/lib/python3.12/asyncio/base_events.py", line 687, in run_until_complete
return future.result()
^^^^^^^^^^^^^^^
File "/mnt/d/repos/llama/llama3/.venv/lib/python3.12/site-packages/llama_stack/cli/download.py", line 263, in download
await self.get_file_info(client)
File "/mnt/d/repos/llama/llama3/.venv/lib/python3.12/site-packages/llama_stack/cli/download.py", line 252, in get_file_info
response.raise_for_status()
File "/mnt/d/repos/llama/llama3/.venv/lib/python3.12/site-packages/httpx/_models.py", line 763, in raise_for_status
raise HTTPStatusError(message, request=request, response=self)
httpx.HTTPStatusError: Client error '405 Forbidden.' for url 'https://llama3-1.llamameta.net/Llama-3.1-70B-Instruct/consolidated.01.pth?Policy=<redacted>'
For more information check: https://developer.mozilla.org/en-US/docs/Web/HTTP/Status/405
I want to download 3.1-70B-Instruct using the python llama stack client. When downloading the 8B-Instruct Model it works fine. Other tested models also work fine. 70B-Instruct fails on downloading
consolidated.01.pth
,consolidated
00.pth` runs through. Also see the scrollback.Environment is WSL, 24.04 LTS, 3090 GPU
Steps to reproduce
python3 -m venv .venv && source .venv/bin/activate
,llama download --source meta --model-id Llama3.1-70B-Instruct
, paste url and waitExpected behaviour successful download
Actual behaviour download fails on
consolidated.01.pth
with Error 405Scrollback