github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-06 16:11:34 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #3944] /api/embeddings hangs when prompt is only whitespace #48958

New Issue
Closed
opened 2026-04-28 10:17:55 -05:00 by GiteaMirror · 5 comments
GiteaMirror commented 2026-04-28 10:17:55 -05:00
Owner
Copy Link

Originally created by @alexmavr on GitHub (Apr 26, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/3944

Originally assigned to: @jmorganca on GitHub.

What is the issue?

The following invocation hangs indefinitely:

$ curl http://localhost:11434/api/embeddings -d '{
  "model": "all-minilm",
  "prompt": " "
}'

Same behavior for model "mxbai-embed-large"

Relevant debug logs:

{"function":"process_single_task","level":"INFO","line":1510,"msg":"slot data","n_idle_slots":1,"n_processing_slots":0,"task_id":6,"tid":"0x1ea27fac0","timestamp":1714138666}
{"function":"log_server_request","level":"INFO","line":2741,"method":"GET","msg":"request","params":{},"path":"/health","remote_addr":"127.0.0.1","remote_port":53684,"status":200,"tid":"0x16be2b000","timestamp":1714138666}
{"function":"process_single_task","level":"INFO","line":1510,"msg":"slot data","n_idle_slots":1,"n_processing_slots":0,"task_id":7,"tid":"0x1ea27fac0","timestamp":1714138666}
{"function":"log_server_request","level":"INFO","line":2741,"method":"GET","msg":"request","params":{},"path":"/health","remote_addr":"127.0.0.1","remote_port":53689,"status":200,"tid":"0x16beb7000","timestamp":1714138666}
{"function":"launch_slot_with_data","level":"INFO","line":833,"msg":"slot is processing task","slot_id":0,"task_id":8,"tid":"0x1ea27fac0","timestamp":1714138666}
{"function":"update_slots","level":"INFO","line":1840,"msg":"kv cache rm [p0, end)","p0":0,"slot_id":0,"task_id":8,"tid":"0x1ea27fac0","timestamp":1714138666}

OS

macOS

GPU

Apple

CPU

Apple

Ollama version

0.1.32

Originally created by @alexmavr on GitHub (Apr 26, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/3944 Originally assigned to: @jmorganca on GitHub. ### What is the issue? The following invocation hangs indefinitely: ``` $ curl http://localhost:11434/api/embeddings -d '{ "model": "all-minilm", "prompt": " " }' ``` Same behavior for model "mxbai-embed-large" Relevant debug logs: ``` {"function":"process_single_task","level":"INFO","line":1510,"msg":"slot data","n_idle_slots":1,"n_processing_slots":0,"task_id":6,"tid":"0x1ea27fac0","timestamp":1714138666} {"function":"log_server_request","level":"INFO","line":2741,"method":"GET","msg":"request","params":{},"path":"/health","remote_addr":"127.0.0.1","remote_port":53684,"status":200,"tid":"0x16be2b000","timestamp":1714138666} {"function":"process_single_task","level":"INFO","line":1510,"msg":"slot data","n_idle_slots":1,"n_processing_slots":0,"task_id":7,"tid":"0x1ea27fac0","timestamp":1714138666} {"function":"log_server_request","level":"INFO","line":2741,"method":"GET","msg":"request","params":{},"path":"/health","remote_addr":"127.0.0.1","remote_port":53689,"status":200,"tid":"0x16beb7000","timestamp":1714138666} {"function":"launch_slot_with_data","level":"INFO","line":833,"msg":"slot is processing task","slot_id":0,"task_id":8,"tid":"0x1ea27fac0","timestamp":1714138666} {"function":"update_slots","level":"INFO","line":1840,"msg":"kv cache rm [p0, end)","p0":0,"slot_id":0,"task_id":8,"tid":"0x1ea27fac0","timestamp":1714138666} ``` ### OS macOS ### GPU Apple ### CPU Apple ### Ollama version 0.1.32
GiteaMirror added the bug label 2026-04-28 10:17:55 -05:00
GiteaMirror commented 2026-04-28 10:18:00 -05:00
Author
Owner
Copy Link

@brycereitano commented on GitHub (Apr 26, 2024):

I was able to reproduce when using an embedding only model on CPU. When using a chat model it handled it gracefully and produced an embedding. In contrast to an empty string that returns no vectors.

I got the following error when force cancelling the request, which points to a llama.cpp issue instead of a bug with ollama.

time=2024-04-26T07:43:23.124-06:00 level=INFO source=routes.go:485 msg="embedding generation failed: do embedding request: Post \"http://127.0.0.1:3
6759/embedding\": EOF"
<!-- gh-comment-id:2079435193 --> @brycereitano commented on GitHub (Apr 26, 2024): I was able to reproduce when using an embedding only model on CPU. When using a chat model it handled it gracefully and produced an embedding. In contrast to an empty string that returns no vectors. I got the following error when force cancelling the request, which points to a llama.cpp issue instead of a bug with ollama. ``` time=2024-04-26T07:43:23.124-06:00 level=INFO source=routes.go:485 msg="embedding generation failed: do embedding request: Post \"http://127.0.0.1:3 6759/embedding\": EOF" ```
GiteaMirror commented 2026-04-28 10:18:02 -05:00
Author
Owner
Copy Link

@EnGassa commented on GitHub (Apr 26, 2024):

+1

<!-- gh-comment-id:2079436550 --> @EnGassa commented on GitHub (Apr 26, 2024): +1
GiteaMirror commented 2026-04-28 10:18:02 -05:00
Author
Owner
Copy Link

@alexmavr commented on GitHub (Apr 26, 2024):

For more context, the issue initially presented while trying to get the embedding of a string representation including only the UTF-8 BOM byte sequence as rendered by golang with fmt.Sprintf("%s", bomString), so this is not exclusive to whitespace.

<!-- gh-comment-id:2079485289 --> @alexmavr commented on GitHub (Apr 26, 2024): For more context, the issue initially presented while trying to get the embedding of a string representation including only the UTF-8 BOM byte sequence as rendered by golang with `fmt.Sprintf("%s", bomString)`, so this is not exclusive to whitespace.
GiteaMirror commented 2026-04-28 10:18:03 -05:00
Author
Owner
Copy Link

@JornWildt commented on GitHub (May 29, 2024):

Also seen here (Windows, GEForce RTX 4060).

Ollama hangs completely when sending this:

POST {{LocalAddress}}/api/embeddings
Content-Type: application/json
Accept: application/json

{
  "model": "nomic-embed-text",
  "prompt": "\r\n"
}
<!-- gh-comment-id:2136648809 --> @JornWildt commented on GitHub (May 29, 2024): Also seen here (Windows, GEForce RTX 4060). Ollama hangs completely when sending this: ``` POST {{LocalAddress}}/api/embeddings Content-Type: application/json Accept: application/json { "model": "nomic-embed-text", "prompt": "\r\n" } ```
GiteaMirror commented 2026-04-28 10:18:04 -05:00
Author
Owner
Copy Link

@jmorganca commented on GitHub (Jun 29, 2024):

Hi all this should be fixed now - thanks for reporting @alexmavr 😊

<!-- gh-comment-id:2198366534 --> @jmorganca commented on GitHub (Jun 29, 2024): Hi all this should be fixed now - thanks for reporting @alexmavr 😊
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#48958
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?