[GH-ISSUE #4434] Behind PROXY, how to set a proxy in Ollama #2771

Closed
opened 2026-04-12 13:05:29 -05:00 by GiteaMirror · 7 comments
Owner

Originally created by @mchopra80 on GitHub (May 14, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/4434

What is the issue?

I installed Ollama on a Windows Server and on cmd I can call Ollama but when trying to pull a model I get this error:
pulling manifest
Error: pull model manifest: Get https://registry.ollama.ai/v2/library/llama3/manifests/latest: dial tcp: lookup registry.ollama.ai: no such host

I am behind a corporate firewall and need to set the proxy - how do I set the PROXY information in ollama Windows Version? thanks.

OS

Windows

GPU

Other

CPU

Intel

Ollama version

0.1.37

Originally created by @mchopra80 on GitHub (May 14, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/4434 ### What is the issue? I installed Ollama on a Windows Server and on cmd I can call Ollama but when trying to pull a model I get this error: pulling manifest Error: pull model manifest: Get https://registry.ollama.ai/v2/library/llama3/manifests/latest: dial tcp: lookup registry.ollama.ai: no such host I am behind a corporate firewall and need to set the proxy - how do I set the PROXY information in ollama Windows Version? thanks. ### OS Windows ### GPU Other ### CPU Intel ### Ollama version 0.1.37
GiteaMirror added the bug label 2026-04-12 13:05:29 -05:00
Author
Owner

@pdevine commented on GitHub (May 14, 2024):

This is covered in the FAQ. Also check out [how to set environment variables(https://github.com/ollama/ollama/blob/main/docs/faq.md#how-do-i-configure-ollama-server) in the FAQ as well.

Hopefully this helps!

<!-- gh-comment-id:2110884113 --> @pdevine commented on GitHub (May 14, 2024): This is covered in the [FAQ](https://github.com/ollama/ollama/blob/main/docs/faq.md#how-do-i-use-ollama-behind-a-proxy). Also check out [how to set environment variables(https://github.com/ollama/ollama/blob/main/docs/faq.md#how-do-i-configure-ollama-server) in the FAQ as well. Hopefully this helps!
Author
Owner

@WeiguangHan commented on GitHub (Sep 12, 2024):

@mchopra80 Hi, have you solved this problem?

<!-- gh-comment-id:2345152074 --> @WeiguangHan commented on GitHub (Sep 12, 2024): @mchopra80 Hi, have you solved this problem?
Author
Owner

@yifan0011 commented on GitHub (Jan 22, 2025):

@mchopra80 Hi, have you solved this problem?

I didn't, after setting the system variable, I got another error pulling model: Error: max retries exceeded: EOF

<!-- gh-comment-id:2606843469 --> @yifan0011 commented on GitHub (Jan 22, 2025): > [@mchopra80](https://github.com/mchopra80) Hi, have you solved this problem? I didn't, after setting the system variable, I got another error pulling model: Error: max retries exceeded: EOF
Author
Owner

@yangjianlin-c commented on GitHub (Feb 8, 2025):

it can be solved by add environment variable HTTPS_PROXY on windowns.

If still not clear about it, ref to this guide

https://esp32cube.com/post/set-proxy-for-ollma-on-windows/

<!-- gh-comment-id:2644735377 --> @yangjianlin-c commented on GitHub (Feb 8, 2025): it can be solved by add environment variable HTTPS_PROXY on windowns. If still not clear about it, ref to this guide [https://esp32cube.com/post/set-proxy-for-ollma-on-windows/](https://esp32cube.com/post/set-proxy-for-ollma-on-windows/)
Author
Owner

@5yh commented on GitHub (Feb 13, 2025):

you need to reboot ollama after you set HTTPS_PROXY

<!-- gh-comment-id:2655647248 --> @5yh commented on GitHub (Feb 13, 2025): you need to reboot ollama after you set HTTPS_PROXY
Author
Owner

@gavinesl commented on GitHub (Feb 17, 2025):

you need to reboot ollama after you set HTTPS_PROXY

Helped! On windows, run "Quit ollama" after you set HTTPS_PROXY.

<!-- gh-comment-id:2661778298 --> @gavinesl commented on GitHub (Feb 17, 2025): > you need to reboot ollama after you set HTTPS_PROXY Helped! On windows, run "Quit ollama" after you set HTTPS_PROXY.
Author
Owner

@dv0ps commented on GitHub (Mar 31, 2026):

There are other issues at play in addition to setting the environment variable correctly on a windows client.
This particular error "pull model manifest: Get "https://registry.ollama.ai/v2/library//manifests/2b": EOF" can also be related to a corporate firewall proxy with TLS snooping (inspection) security enabled which can cause TLS handshake problems and other TLS connection issues causing the ollama client to error out.

With ollama being a trusted web service many corporate IT dept may allow an exception be made for the ollama's download host: "registry.ollama.ai" Ask your IT group to exclude it from TLS inspection.

<!-- gh-comment-id:4165985151 --> @dv0ps commented on GitHub (Mar 31, 2026): There are other issues at play in addition to setting the environment variable correctly on a windows client. This particular error "pull model manifest: Get "https://registry.ollama.ai/v2/library/<LLMname>/manifests/2b": EOF" can also be related to a corporate firewall proxy with TLS snooping (inspection) security enabled which can cause TLS handshake problems and other TLS connection issues causing the ollama client to error out. With ollama being a trusted web service many corporate IT dept may allow an exception be made for the ollama's download host: "registry.ollama.ai" Ask your IT group to exclude it from TLS inspection.
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/ollama#2771