github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-08 00:51:34 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #15661] pull from hf.co/ fails: realm host huggingface.co does not match original host hf.co #72049

New Issue
Open
opened 2026-05-05 03:24:01 -05:00 by GiteaMirror · 1 comment
GiteaMirror commented 2026-05-05 03:24:01 -05:00
Owner
Copy Link

Originally created by @dasoliver on GitHub (Apr 18, 2026).
Original GitHub issue: https://github.com/ollama/ollama/issues/15661

Ollama Issue Draft — hf.co registry host mismatch

Title

Pull from hf.co/... fails with "realm host huggingface.co does not match original host hf.co"

Body

Environment

  • Ollama version: (run ollama --version)
  • OS: macOS 26 (Apple Silicon M3 Max)
  • Network: Starlink

Command

ollama pull hf.co/ggerganov/whisper.cpp-large-v3:q8_0

Expected

The manifest is fetched and the model is pulled like any other HuggingFace-hosted GGUF.

Actual

pulling manifest
Error: pull model manifest: realm host "huggingface.co" does not match original host "hf.co"

Analysis

The Ollama client accepts hf.co/* as a shorthand for HuggingFace. When the backend redirects the manifest realm header to huggingface.co, the client refuses because the realm host does not match the originally-requested host. The resolver should treat hf.co and huggingface.co as the same registry, since hf.co is the official short alias.

Workarounds

  • Install whisper.cpp natively via brew install whisper-cpp (not an Ollama model, different runtime).
  • Try the fully qualified huggingface.co/<owner>/<model> — reportedly sometimes works.

Suggested Fix

In the registry client (likely server/images.go or registry/*.go), add hf.co and huggingface.co to an equivalence set so the post-auth realm host check does not trip.

References

  • Related: issue #10500 (HF 401 Invalid credentials)
  • Related: issue #13297 (HF token asked despite no auth)
  • Related: issue #13768 (HF models incorrectly marked no-tools)
Originally created by @dasoliver on GitHub (Apr 18, 2026). Original GitHub issue: https://github.com/ollama/ollama/issues/15661 # Ollama Issue Draft — hf.co registry host mismatch ## Title Pull from `hf.co/...` fails with "realm host huggingface.co does not match original host hf.co" ## Body ### Environment - Ollama version: (run `ollama --version`) - OS: macOS 26 (Apple Silicon M3 Max) - Network: Starlink ### Command ```bash ollama pull hf.co/ggerganov/whisper.cpp-large-v3:q8_0 ``` ### Expected The manifest is fetched and the model is pulled like any other HuggingFace-hosted GGUF. ### Actual ``` pulling manifest Error: pull model manifest: realm host "huggingface.co" does not match original host "hf.co" ``` ### Analysis The Ollama client accepts `hf.co/*` as a shorthand for HuggingFace. When the backend redirects the manifest realm header to `huggingface.co`, the client refuses because the realm host does not match the originally-requested host. The resolver should treat `hf.co` and `huggingface.co` as the same registry, since `hf.co` is the official short alias. ### Workarounds - Install whisper.cpp natively via `brew install whisper-cpp` (not an Ollama model, different runtime). - Try the fully qualified `huggingface.co/<owner>/<model>` — reportedly sometimes works. ### Suggested Fix In the registry client (likely `server/images.go` or `registry/*.go`), add `hf.co` and `huggingface.co` to an equivalence set so the post-auth realm host check does not trip. ### References - Related: issue #10500 (HF 401 Invalid credentials) - Related: issue #13297 (HF token asked despite no auth) - Related: issue #13768 (HF models incorrectly marked no-tools)
GiteaMirror commented 2026-05-05 03:24:02 -05:00
Author
Owner
Copy Link

@PureBlissAK commented on GitHub (Apr 18, 2026):

🤖 Automated Triage & Analysis Report

Issue: #15661
Analyzed: 2026-04-18T18:13:48.175046

Analysis

  • Type: unknown
  • Severity: medium
  • Components: unknown

Implementation Plan

  • Effort: medium
  • Steps:

This issue has been triaged and marked for implementation.

<!-- gh-comment-id:4274294958 --> @PureBlissAK commented on GitHub (Apr 18, 2026): <!-- ollama-issue-orchestrator:v1 issue:15661 --> ## 🤖 Automated Triage & Analysis Report **Issue**: #15661 **Analyzed**: 2026-04-18T18:13:48.175046 ### Analysis - **Type**: unknown - **Severity**: medium - **Components**: unknown ### Implementation Plan - **Effort**: medium - **Steps**: *This issue has been triaged and marked for implementation.*
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
amd
api
app
bug
build
cli
cloud
compatibility
context-length
create
docker
documentation
embeddings
feature request
feedback wanted
good first issue
gpt-oss
gpu
harmony
help wanted
image
install
intel
js
launch
linux
macos
memory
mlx
model
needs more info
networking
nvidia
ollama.com
performance
pull-request
Mirrored from GitHub Pull Request
 python
question
registry
rendering
thinking
tools
top
vulkan
windows
wsl
No Label
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
GiteaMirror ninjasurge
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/ollama#72049
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?