github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-05 23:53:43 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #12633] Qwen3-embedding:0.6b does not work on v0.12.5 , but works on previous versions #8385

New Issue
Closed
opened 2026-04-12 21:01:57 -05:00 by GiteaMirror · 2 comments
GiteaMirror commented 2026-04-12 21:01:57 -05:00
Owner
Copy Link

Originally created by @Kemo-is-kiwi on GitHub (Oct 15, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/12633

What is the issue?

Issue

qwen3-embedding:0.6b model crashes immediately when using /api/embed on Ollama v0.12.5 (Windows, CPU backend).

Worked correctly on previous versions (e.g. v0.12.2 / v0.12.3).

Environment

Ollama version: v0.12.5
OS: Windows 11 (x64)
Hardware: CPU only (no GPU detected)
Model: qwen3-embedding:0.6b
Command: POST /api/embed

Repro Steps

ollama pull qwen3-embedding:0.6b
ollama serve

Then in Python:

import requests payload = {"model":"qwen3-embedding:0.6b","input":"t"} print(requests.post("http://127.0.0.1:11434/api/embed", json=payload).json())

Relevant log output

Observed Log Output

print_info: model type = 0.6B
print_info: model params = 595.78 M
print_info: general.name = Qwen3 Embedding 0.6b
print_info: n_ctx_orig_yarn = 32768
print_info: rope type = 2
print_info: rope scaling = linear
print_info: n_ctx_per_seq (4096) < n_ctx_train (32768) -- the full capacity of the model will not be utilized

llama_context: CPU compute buffer size = 1104.01 MiB
llama_context: graph nodes = 1127
llama_context: graph splits = 1

time=2025-10-15T04:20:21.574+03:00 level=INFO source=server.go:1309 msg="llama runner started in 1.15 seconds"
time=2025-10-15T04:20:21.575+03:00 level=DEBUG source=sched.go:493 msg="finished setting up" runner.name=registry.ollama.ai/library/qwen3-embedding:0.6b runner.vram="0 B" runner.model=C:\Users\user\.ollama\models\blobs\sha256-06507c7b42688469c4e7298b0a1e16deff06caf291cf0a5b278c308249c3e439 runner.num_ctx=42000

C:/a/ollama/ollama/ml/backend/ggml/ggml/src/ggml-cpu/ops.cpp:4663:
GGML_ASSERT(i01 >= 0 && i01 < ne01) failed

[GIN] 2025/10/15 - 04:20:21 | 500 | 1.8646282s | 127.0.0.1 | POST "/api/embed"

time=2025-10-15T04:20:22.045+03:00 level=ERROR source=server.go:426 msg="llama runner terminated" error="exit status 0xc0000409"


---

Error from Python client

ollama._types.ResponseError: do embedding request: Post "http://127.0.0.1:63883/embedding":
read tcp 127.0.0.1:63899->127.0.0.1:63883: wsarecv:
An existing connection was forcibly closed by the remote host. (status code: 500)

OS

Windows

GPU

Other

CPU

Intel

Ollama version

v0.12.5

Originally created by @Kemo-is-kiwi on GitHub (Oct 15, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/12633 ### What is the issue? Issue qwen3-embedding:0.6b model crashes immediately when using /api/embed on Ollama v0.12.5 (Windows, CPU backend). Worked correctly on previous versions (e.g. v0.12.2 / v0.12.3). Environment Ollama version: v0.12.5 OS: Windows 11 (x64) Hardware: CPU only (no GPU detected) Model: qwen3-embedding:0.6b Command: POST /api/embed --- Repro Steps ollama pull qwen3-embedding:0.6b ollama serve Then in Python: import requests payload = {"model":"qwen3-embedding:0.6b","input":"t"} print(requests.post("http://127.0.0.1:11434/api/embed", json=payload).json()) ### Relevant log output ```shell Observed Log Output print_info: model type = 0.6B print_info: model params = 595.78 M print_info: general.name = Qwen3 Embedding 0.6b print_info: n_ctx_orig_yarn = 32768 print_info: rope type = 2 print_info: rope scaling = linear print_info: n_ctx_per_seq (4096) < n_ctx_train (32768) -- the full capacity of the model will not be utilized llama_context: CPU compute buffer size = 1104.01 MiB llama_context: graph nodes = 1127 llama_context: graph splits = 1 time=2025-10-15T04:20:21.574+03:00 level=INFO source=server.go:1309 msg="llama runner started in 1.15 seconds" time=2025-10-15T04:20:21.575+03:00 level=DEBUG source=sched.go:493 msg="finished setting up" runner.name=registry.ollama.ai/library/qwen3-embedding:0.6b runner.vram="0 B" runner.model=C:\Users\user\.ollama\models\blobs\sha256-06507c7b42688469c4e7298b0a1e16deff06caf291cf0a5b278c308249c3e439 runner.num_ctx=42000 C:/a/ollama/ollama/ml/backend/ggml/ggml/src/ggml-cpu/ops.cpp:4663: GGML_ASSERT(i01 >= 0 && i01 < ne01) failed [GIN] 2025/10/15 - 04:20:21 | 500 | 1.8646282s | 127.0.0.1 | POST "/api/embed" time=2025-10-15T04:20:22.045+03:00 level=ERROR source=server.go:426 msg="llama runner terminated" error="exit status 0xc0000409" --- Error from Python client ollama._types.ResponseError: do embedding request: Post "http://127.0.0.1:63883/embedding": read tcp 127.0.0.1:63899->127.0.0.1:63883: wsarecv: An existing connection was forcibly closed by the remote host. (status code: 500) ``` ### OS Windows ### GPU Other ### CPU Intel ### Ollama version v0.12.5
GiteaMirror added the bug label 2026-04-12 21:01:57 -05:00
GiteaMirror commented 2026-04-12 21:01:59 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Oct 15, 2025):

#12014

<!-- gh-comment-id:3405852518 --> @rick-github commented on GitHub (Oct 15, 2025): #12014
GiteaMirror commented 2026-04-12 21:02:00 -05:00
Author
Owner
Copy Link

@jmorganca commented on GitHub (Oct 17, 2025):

This should be fixed in 0.12.6. I'm so sorry for the issue in 0.12.5. Let me know if you're still seeing issues

<!-- gh-comment-id:3416713933 --> @jmorganca commented on GitHub (Oct 17, 2025): This should be fixed in 0.12.6. I'm so sorry for the issue in 0.12.5. Let me know if you're still seeing issues
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#8385
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?