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 #8520] $OLLAMA_MODELS no longer respected? #52004

New Issue
Closed
opened 2026-04-28 21:36:28 -05:00 by GiteaMirror · 4 comments
GiteaMirror commented 2026-04-28 21:36:28 -05:00
Owner
Copy Link

Originally created by @yuimbo on GitHub (Jan 21, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/8520

What is the issue?

I've been using the OLLAMA_MODELS variable to store my models on an external drive.

I can see that this is set:

$ echo $OLLAMA_MODELS

/Volumes/bigdrive/ollama/models

I can see that my models are stored there:

$ tree $OLLAMA_MODELS

/Volumes/bigdrive/ollama/models
├── blobs
│   ├── sha256-0b4284c1f87029e67654c7953afa16279961632cf73dcfe33374c4c2f298fa35
│   ├── sha256-1ae29500b4be5bb4ce3981e3692ee8689ce5df0ae3080ed4c5ff9f72bf01ba6a
│   ├── sha256-2eedb02591412148c2fea86b5896da88ec5cddea551bcccde3270aa9d1f048ff
│   ....
└── manifests
    └── registry.ollama.ai
        ├── alibayram
        │   └── erurollm-9b-instruct
        │       └── latest
        ├── huihui_ai
        │   ├── llama3.3-abliterated
        │   │   └── latest
        │   └── qwq-abliterated
        │       └── latest
        ├── library
        │   ├── deepseek-r1
        │   │   ├── 32b
        │   │   └── 70b
        ....

But for some reason the models does not show up in ollama?

$ ollama list

NAME               ID              SIZE     MODIFIED

It seems the OLLAMA_MODELS variable is not being respected.

If there are any debug flags I can use to inspect what's going wrong, let me know!

OS

macOS

GPU

Apple

CPU

Apple

Ollama version

0.5.7

Originally created by @yuimbo on GitHub (Jan 21, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/8520 ### What is the issue? I've been using the OLLAMA_MODELS variable to store my models on an external drive. I can see that this is set: ``` $ echo $OLLAMA_MODELS /Volumes/bigdrive/ollama/models ``` I can see that my models are stored there: ``` $ tree $OLLAMA_MODELS /Volumes/bigdrive/ollama/models ├── blobs │   ├── sha256-0b4284c1f87029e67654c7953afa16279961632cf73dcfe33374c4c2f298fa35 │   ├── sha256-1ae29500b4be5bb4ce3981e3692ee8689ce5df0ae3080ed4c5ff9f72bf01ba6a │   ├── sha256-2eedb02591412148c2fea86b5896da88ec5cddea551bcccde3270aa9d1f048ff │   .... └── manifests └── registry.ollama.ai ├── alibayram │   └── erurollm-9b-instruct │   └── latest ├── huihui_ai │   ├── llama3.3-abliterated │   │   └── latest │   └── qwq-abliterated │   └── latest ├── library │   ├── deepseek-r1 │   │   ├── 32b │   │   └── 70b .... ``` But for some reason the models does not show up in ollama? ``` $ ollama list NAME ID SIZE MODIFIED ``` It seems the OLLAMA_MODELS variable is not being respected. If there are any debug flags I can use to inspect what's going wrong, let me know! ### OS macOS ### GPU Apple ### CPU Apple ### Ollama version 0.5.7
GiteaMirror added the bug label 2026-04-28 21:36:28 -05:00
GiteaMirror commented 2026-04-28 21:36:30 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Jan 21, 2025):

Server logs.

<!-- gh-comment-id:2605181680 --> @rick-github commented on GitHub (Jan 21, 2025): [Server logs](https://github.com/ollama/ollama/blob/main/docs/troubleshooting.md#how-to-troubleshoot-issues).
GiteaMirror commented 2026-04-28 21:36:30 -05:00
Author
Owner
Copy Link

@yuimbo commented on GitHub (Jan 21, 2025):

2025/01/21 16:49:41 routes.go:1187: INFO server config env="map[HTTPS_PROXY: HTTP_PROXY: NO_PROXY: OLLAMA_DEBUG:false OLLAMA_FLASH_ATTENTION:false OLLAMA_GPU_OVERHEAD:0 OLLAMA_HOST:http://127.0.0.1:11434 OLLAMA_KEEP_ALIVE:5m0s OLLAMA_KV_CACHE_TYPE: OLLAMA_LLM_LIBRARY: OLLAMA_LOAD_TIMEOUT:5m0s OLLAMA_MAX_LOADED_MODELS:0 OLLAMA_MAX_QUEUE:512 OLLAMA_MODELS:/Users/username/.ollama/models OLLAMA_MULTIUSER_CACHE:false OLLAMA_NOHISTORY:false OLLAMA_NOPRUNE:false OLLAMA_NUM_PARALLEL:0 OLLAMA_ORIGINS:[http://localhost https://localhost http://localhost:* https://localhost:* http://127.0.0.1 https://127.0.0.1 http://127.0.0.1:* https://127.0.0.1:* http://0.0.0.0 https://0.0.0.0 http://0.0.0.0:* https://0.0.0.0:* app://* file://* tauri://* vscode-webview://*] OLLAMA_SCHED_SPREAD:false http_proxy: https_proxy: no_proxy:]"

I can see in server log that it seems to resolve to the default folder instead OLLAMA_MODELS:/Users/username/.ollama/models.

Possibly related to this earlier:

2025/01/08 16:06:48 routes.go:1259: INFO server config env="map[HTTPS_PROXY: HTTP_PROXY: NO_PROXY: OLLAMA_DEBUG:false OLLAMA_FLASH_ATTENTION:false OLLAMA_GPU_OVERHEAD:0 OLLAMA_HOST:http://127.0.0.1:11434 OLLAMA_KEEP_ALIVE:5m0s OLLAMA_KV_CACHE_TYPE: OLLAMA_LLM_LIBRARY: OLLAMA_LOAD_TIMEOUT:5m0s OLLAMA_MAX_LOADED_MODELS:0 OLLAMA_MAX_QUEUE:512 OLLAMA_MODELS:/Volumes/bigdrive/ollama/models OLLAMA_MULTIUSER_CACHE:false OLLAMA_NOHISTORY:false OLLAMA_NOPRUNE:false OLLAMA_NUM_PARALLEL:0 OLLAMA_ORIGINS:[http://localhost https://localhost http://localhost:* https://localhost:* http://127.0.0.1 https://127.0.0.1 http://127.0.0.1:* https://127.0.0.1:* http://0.0.0.0 https://0.0.0.0 http://0.0.0.0:* https://0.0.0.0:* app://* file://* tauri://* vscode-webview://*] OLLAMA_SCHED_SPREAD:false http_proxy: https_proxy: no_proxy:]"
Error: mkdir /Volumes/bigdrive: permission denied

Looks like it couldn't access the drive for a while (probably when I didn't have it plugged in).

Has my models directory been marked as dirty/inaccessible? Can I reset this somehow?

Though this might be a red herring, I still have a lot of logs of the server loading models from the external drive after that point in time.

<!-- gh-comment-id:2605223874 --> @yuimbo commented on GitHub (Jan 21, 2025): ``` 2025/01/21 16:49:41 routes.go:1187: INFO server config env="map[HTTPS_PROXY: HTTP_PROXY: NO_PROXY: OLLAMA_DEBUG:false OLLAMA_FLASH_ATTENTION:false OLLAMA_GPU_OVERHEAD:0 OLLAMA_HOST:http://127.0.0.1:11434 OLLAMA_KEEP_ALIVE:5m0s OLLAMA_KV_CACHE_TYPE: OLLAMA_LLM_LIBRARY: OLLAMA_LOAD_TIMEOUT:5m0s OLLAMA_MAX_LOADED_MODELS:0 OLLAMA_MAX_QUEUE:512 OLLAMA_MODELS:/Users/username/.ollama/models OLLAMA_MULTIUSER_CACHE:false OLLAMA_NOHISTORY:false OLLAMA_NOPRUNE:false OLLAMA_NUM_PARALLEL:0 OLLAMA_ORIGINS:[http://localhost https://localhost http://localhost:* https://localhost:* http://127.0.0.1 https://127.0.0.1 http://127.0.0.1:* https://127.0.0.1:* http://0.0.0.0 https://0.0.0.0 http://0.0.0.0:* https://0.0.0.0:* app://* file://* tauri://* vscode-webview://*] OLLAMA_SCHED_SPREAD:false http_proxy: https_proxy: no_proxy:]" ``` I can see in server log that it seems to resolve to the default folder instead `OLLAMA_MODELS:/Users/username/.ollama/models`. Possibly related to this earlier: ``` 2025/01/08 16:06:48 routes.go:1259: INFO server config env="map[HTTPS_PROXY: HTTP_PROXY: NO_PROXY: OLLAMA_DEBUG:false OLLAMA_FLASH_ATTENTION:false OLLAMA_GPU_OVERHEAD:0 OLLAMA_HOST:http://127.0.0.1:11434 OLLAMA_KEEP_ALIVE:5m0s OLLAMA_KV_CACHE_TYPE: OLLAMA_LLM_LIBRARY: OLLAMA_LOAD_TIMEOUT:5m0s OLLAMA_MAX_LOADED_MODELS:0 OLLAMA_MAX_QUEUE:512 OLLAMA_MODELS:/Volumes/bigdrive/ollama/models OLLAMA_MULTIUSER_CACHE:false OLLAMA_NOHISTORY:false OLLAMA_NOPRUNE:false OLLAMA_NUM_PARALLEL:0 OLLAMA_ORIGINS:[http://localhost https://localhost http://localhost:* https://localhost:* http://127.0.0.1 https://127.0.0.1 http://127.0.0.1:* https://127.0.0.1:* http://0.0.0.0 https://0.0.0.0 http://0.0.0.0:* https://0.0.0.0:* app://* file://* tauri://* vscode-webview://*] OLLAMA_SCHED_SPREAD:false http_proxy: https_proxy: no_proxy:]" Error: mkdir /Volumes/bigdrive: permission denied ``` Looks like it couldn't access the drive for a while (probably when I didn't have it plugged in). Has my models directory been marked as dirty/inaccessible? Can I reset this somehow? Though this might be a red herring, I still have a lot of logs of the server loading models from the external drive after that point in time.
GiteaMirror commented 2026-04-28 21:36:31 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Jan 21, 2025):

Are you sure OLLAMA_MODELS is set in the launch context? What does the following show:

launchctl getenv OLLAMA_MODELS
<!-- gh-comment-id:2605243640 --> @rick-github commented on GitHub (Jan 21, 2025): Are you sure OLLAMA_MODELS is set in the launch context? What does the following show: ```sh launchctl getenv OLLAMA_MODELS ```
GiteaMirror commented 2026-04-28 21:36:31 -05:00
Author
Owner
Copy Link

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

Ah yes you are right rick, I had just naively set the variable in my .zshrc 🙃

<!-- gh-comment-id:2606724011 --> @yuimbo commented on GitHub (Jan 22, 2025): Ah yes you are right rick, I had just naively set the variable in my .zshrc 🙃
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 bug
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#52004
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?