github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-07 00:22:43 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #7490] Return an empty embed list #66820

New Issue
Closed
opened 2026-05-04 08:16:49 -05:00 by GiteaMirror · 3 comments
GiteaMirror commented 2026-05-04 08:16:49 -05:00
Owner
Copy Link

Originally created by @somnifex on GitHub (Nov 4, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/7490

What is the issue?

When I use the bge-m3 model, the return is an empty list regardless of the input.
My test code:

import requests
url = "http://localhost:12121/api/embeddings"
payload = {
    "model": "bge-m3",
    "input": "lol"
}
response = requests.post(url, json=payload)
print(response.text)

run program output

{"embedding":[]}

OS

Docker

GPU

Nvidia

CPU

Intel

Ollama version

0.3.14

Originally created by @somnifex on GitHub (Nov 4, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/7490 ### What is the issue? When I use the [bge-m3](https://ollama.com/library/bge-m3) model, the return is an empty list regardless of the input. My test code: ```python import requests url = "http://localhost:12121/api/embeddings" payload = { "model": "bge-m3", "input": "lol" } response = requests.post(url, json=payload) print(response.text) ``` run program output ``` {"embedding":[]} ``` ### OS Docker ### GPU Nvidia ### CPU Intel ### Ollama version 0.3.14
GiteaMirror added the bug label 2026-05-04 08:16:49 -05:00
GiteaMirror commented 2026-05-04 08:16:53 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Nov 4, 2024):

/api/embeddings is deprecated, use /api/embed. The reason the returned embedding is empty is because /api/embeddings uses prompt not input for the text to process.

<!-- gh-comment-id:2454786332 --> @rick-github commented on GitHub (Nov 4, 2024): `/api/embeddings` is [deprecated](https://github.com/ollama/ollama/blob/main/docs/api.md#generate-embedding), use `/api/embed`. The reason the returned embedding is empty is because `/api/embeddings` uses `prompt` not `input` for the text to process.
GiteaMirror commented 2026-05-04 08:16:54 -05:00
Author
Owner
Copy Link

@somnifex commented on GitHub (Nov 4, 2024):

/api/embeddings is deprecated, use /api/embed. The reason the returned embedding is empty is because /api/embeddings uses prompt not input for the text to process.

thanks

<!-- gh-comment-id:2454790553 --> @somnifex commented on GitHub (Nov 4, 2024): > `/api/embeddings` is [deprecated](https://github.com/ollama/ollama/blob/main/docs/api.md#generate-embedding), use `/api/embed`. The reason the returned embedding is empty is because `/api/embeddings` uses `prompt` not `input` for the text to process. thanks
GiteaMirror commented 2026-05-04 08:16:56 -05:00
Author
Owner
Copy Link

@lucid-newt commented on GitHub (Jan 16, 2025):

I had the same issue. /api/embeddings seems to be depreciated, but input is necessary to get embeddings (using prompt returns an empty embedding). At least with Ollama API Version 0.5.4 on Windows 11.

Here is a test script to help anyone with troubleshooting

'''
January 16, 2025
This script tests the Ollama API endpoints for the embedding models.
* Generated with GPT-4.

https://github.com/ollama/ollama/issues/7490
https://github.com/ollama/ollama/blob/main/docs/troubleshooting.md
https://github.com/ollama/ollama/blob/main/docs/api.md#generate-embeddings
'''
import requests
import pprint

EMBEDDING_MODEL: str = "nomic-embed-text:latest"

def get_ollama_api_version():
    url = "http://localhost:11434/api/version"
    try:
        response = requests.get(url, timeout=10)
        if response.status_code == 200:
            data = response.json()
            version = data.get("version", "Version not found")
            print(f"API version: {version}\n")
        else:
            print(f"Failed to get API version. Status code: {response.status_code}")
            return None
    except requests.exceptions.RequestException as e:
        print(f"Error connecting to the version endpoint: {e}")
        return None


def get_ollama_running_models():
    """
    Test the /api/ps endpoint at http://localhost:11434/api/ps.
    """
    url = "http://localhost:11434/api/ps"
    headers = {
        "Content-Type": "application/json"
    }

    try:
        response = requests.get(url, headers=headers, timeout=10)
        if response.status_code == 200:
            data = response.json()
            print("PS response:")
            pprint.pprint(data)
        else:
            print(f"Endpoint returned an error. Status code: {response.status_code}")
            print("Response headers:", response.headers)
            print("Response body:", response.text)
    except requests.exceptions.RequestException as e:
        print(f"Error connecting to the endpoint: {e}")


def test_embedding_endpoint():
    """
    Test the embedding endpoint at http://localhost:11434/api/embeddings.
    """
    url = "http://localhost:11434/api/embed"
    payload = {
        "model": EMBEDDING_MODEL,
        "input": "This is a sample text to test embeddings."
    }
    headers = {
        "Content-Type": "application/json"
    }

    try:
        response = requests.post(url, json=payload, headers=headers, timeout=30)
        if response.status_code == 200:
            data = response.json()
            print(f"\nEmbedding response:")
            pprint.pprint(data)
            if not data.get("embeddings"):
                print("Warning: Embedding is empty.")
            else:
                print(f"Embedding generated successfully.")
        else:
            print(f"Endpoint returned an error. Status code: {response.status_code}")
            print("Response headers:", response.headers)
            print("Response body:", response.text)
    except requests.exceptions.RequestException as e:
        print(f"Error connecting to the endpoint: {e}")


if __name__ == "__main__":
    # Test the API version endpoint
    api_version = get_ollama_api_version()

    # Get the running models
    get_ollama_running_models()

    # Test the embedding endpoint
    test_embedding_endpoint()

Example output:

API version: 0.5.4

PS response:
{'models': [{'details': {'families': ['nomic-bert'],
                         'family': 'nomic-bert',
                         'format': 'gguf',
                         'parameter_size': '137M',
                         'parent_model': '',
                         'quantization_level': 'F16'},
             'digest': '0a109f422b47e3a30ba2b10eca18548e944e8a23073ee3f3e947efcf3c45e59f',
             'expires_at': '2025-01-16T01:31:05.7450267-05:00',
             'model': 'nomic-embed-text:latest',
             'name': 'nomic-embed-text:latest',
             'size': 370002944,
             'size_vram': 0}]}

Embedding response:
{'embeddings': [[0.030008234,
                 0.058340646,
                 -0.16748141,
                  ...
                 -0.01700322]],
 'load_duration': 4340300,
 'model': 'nomic-embed-text:latest',
 'prompt_eval_count': 12,
 'total_duration': 21831900}
Embedding generated successfully.
<!-- gh-comment-id:2594642633 --> @lucid-newt commented on GitHub (Jan 16, 2025): I had the same issue. `/api/embeddings` seems to be depreciated, but `input` is necessary to get embeddings (using `prompt` returns an empty embedding). At least with Ollama API Version 0.5.4 on Windows 11. ## Here is a test script to help anyone with troubleshooting ```python ''' January 16, 2025 This script tests the Ollama API endpoints for the embedding models. * Generated with GPT-4. https://github.com/ollama/ollama/issues/7490 https://github.com/ollama/ollama/blob/main/docs/troubleshooting.md https://github.com/ollama/ollama/blob/main/docs/api.md#generate-embeddings ''' import requests import pprint EMBEDDING_MODEL: str = "nomic-embed-text:latest" def get_ollama_api_version(): url = "http://localhost:11434/api/version" try: response = requests.get(url, timeout=10) if response.status_code == 200: data = response.json() version = data.get("version", "Version not found") print(f"API version: {version}\n") else: print(f"Failed to get API version. Status code: {response.status_code}") return None except requests.exceptions.RequestException as e: print(f"Error connecting to the version endpoint: {e}") return None def get_ollama_running_models(): """ Test the /api/ps endpoint at http://localhost:11434/api/ps. """ url = "http://localhost:11434/api/ps" headers = { "Content-Type": "application/json" } try: response = requests.get(url, headers=headers, timeout=10) if response.status_code == 200: data = response.json() print("PS response:") pprint.pprint(data) else: print(f"Endpoint returned an error. Status code: {response.status_code}") print("Response headers:", response.headers) print("Response body:", response.text) except requests.exceptions.RequestException as e: print(f"Error connecting to the endpoint: {e}") def test_embedding_endpoint(): """ Test the embedding endpoint at http://localhost:11434/api/embeddings. """ url = "http://localhost:11434/api/embed" payload = { "model": EMBEDDING_MODEL, "input": "This is a sample text to test embeddings." } headers = { "Content-Type": "application/json" } try: response = requests.post(url, json=payload, headers=headers, timeout=30) if response.status_code == 200: data = response.json() print(f"\nEmbedding response:") pprint.pprint(data) if not data.get("embeddings"): print("Warning: Embedding is empty.") else: print(f"Embedding generated successfully.") else: print(f"Endpoint returned an error. Status code: {response.status_code}") print("Response headers:", response.headers) print("Response body:", response.text) except requests.exceptions.RequestException as e: print(f"Error connecting to the endpoint: {e}") if __name__ == "__main__": # Test the API version endpoint api_version = get_ollama_api_version() # Get the running models get_ollama_running_models() # Test the embedding endpoint test_embedding_endpoint() ``` ### Example output: ``` API version: 0.5.4 PS response: {'models': [{'details': {'families': ['nomic-bert'], 'family': 'nomic-bert', 'format': 'gguf', 'parameter_size': '137M', 'parent_model': '', 'quantization_level': 'F16'}, 'digest': '0a109f422b47e3a30ba2b10eca18548e944e8a23073ee3f3e947efcf3c45e59f', 'expires_at': '2025-01-16T01:31:05.7450267-05:00', 'model': 'nomic-embed-text:latest', 'name': 'nomic-embed-text:latest', 'size': 370002944, 'size_vram': 0}]} Embedding response: {'embeddings': [[0.030008234, 0.058340646, -0.16748141, ... -0.01700322]], 'load_duration': 4340300, 'model': 'nomic-embed-text:latest', 'prompt_eval_count': 12, 'total_duration': 21831900} Embedding generated successfully. ```
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#66820
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?