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 #1707] [Bug] Calling embedding endpoint within short time will receive None #26725

New Issue
Closed
opened 2026-04-22 03:11:54 -05:00 by GiteaMirror · 1 comment
GiteaMirror commented 2026-04-22 03:11:54 -05:00
Owner
Copy Link

Originally created by @samx81 on GitHub (Dec 25, 2023).
Original GitHub issue: https://github.com/ollama/ollama/issues/1707

I'm currently using llama_index to make LLM do a document QA,
but I notice the embedding endpoint often output:

{"embedding":null}

when I do condense_question chat mode which calls LLM to rephrase question (and match context) before actually answer.

Like:

llm = Ollama(model="dolphin2.2-mistral:7b-q4_K_M")
ollama_embedding = OllamaEmbedding(
    model_name="dolphin2.2-mistral:7b-q4_K_M",
    ollama_additional_kwargs={"mirostat": 0},
)

service_context = ServiceContext.from_defaults(llm=llm, embed_model=ollama_embedding)
chat_engine = index.as_chat_engine(service_context=service_context,chat_mode='condense_plus_context')

resp= chat_engine.chat('abc')
print(resp)
# Often happens when LLM is called second time
resp= chat_engine.chat('cde')
print(resp)

I guess it is probably due to ollama or llama.cpp needs to handle different behavior in the same time?
Because if I modify llama_index to check for null and retry, the above code will work:

## In llama_index.embeddings.OllamaEmbedding
while True:
    response = requests.post(
        url=f"{self.base_url}/api/embeddings",
        headers={"Content-Type": "application/json"},
        json=ollama_request_body,
    )
    response.encoding = "utf-8"
    
    if response.status_code != 200:
        optional_detail = response.json().get("error")
        raise ValueError(
            f"Ollama call failed with status code {response.status_code}."
            f" Details: {optional_detail}"
        )
    if response.json()["embedding"]:
        break
Originally created by @samx81 on GitHub (Dec 25, 2023). Original GitHub issue: https://github.com/ollama/ollama/issues/1707 I'm currently using `llama_index` to make LLM do a document QA, but I notice the embedding endpoint often output: ```json {"embedding":null} ``` when I do `condense_question` chat mode which calls LLM to rephrase question (and match context) before actually answer. Like: ```python llm = Ollama(model="dolphin2.2-mistral:7b-q4_K_M") ollama_embedding = OllamaEmbedding( model_name="dolphin2.2-mistral:7b-q4_K_M", ollama_additional_kwargs={"mirostat": 0}, ) service_context = ServiceContext.from_defaults(llm=llm, embed_model=ollama_embedding) chat_engine = index.as_chat_engine(service_context=service_context,chat_mode='condense_plus_context') resp= chat_engine.chat('abc') print(resp) # Often happens when LLM is called second time resp= chat_engine.chat('cde') print(resp) ``` I guess it is probably due to ollama or llama.cpp needs to handle different behavior in the same time? Because if I modify `llama_index` to check for `null` and retry, the above code will work: ```python ## In llama_index.embeddings.OllamaEmbedding while True: response = requests.post( url=f"{self.base_url}/api/embeddings", headers={"Content-Type": "application/json"}, json=ollama_request_body, ) response.encoding = "utf-8" if response.status_code != 200: optional_detail = response.json().get("error") raise ValueError( f"Ollama call failed with status code {response.status_code}." f" Details: {optional_detail}" ) if response.json()["embedding"]: break ```
GiteaMirror added the bug label 2026-04-22 03:11:54 -05:00
GiteaMirror commented 2026-04-22 03:11:57 -05:00
Author
Owner
Copy Link

@jmorganca commented on GitHub (May 6, 2024):

Hi all, this should be fixed now!

<!-- gh-comment-id:2097104278 --> @jmorganca commented on GitHub (May 6, 2024): Hi all, this should be fixed now!
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#26725
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?