github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-07-17 13:47:08 -05:00
Code Issues 16.9k Packages Projects Releases 122 Wiki Activity

[GH-ISSUE #6089] Match behavior of text-generation webui and koboldcpp by accepting requests to v1/completions that don't specify the model. #65843

New Issue
Closed
opened 2026-05-03 22:52:08 -05:00 by GiteaMirror · 4 comments
GiteaMirror commented 2026-05-03 22:52:08 -05:00
Owner
Copy Link

Originally created by @balisujohn on GitHub (Jul 31, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/6089

This works:

url = "http://localhost:11434/v1/completions"

headers = {
    'Content-Type': 'application/json'
}

data = {
    'model':"moondream",
    'prompt': "What is the cat holding?",
    'max_tokens': 20,
    'temperature': 1,
    'top_p': 0.9,
    'seed': 10
}

# Convert data to JSON format
json_data = json.dumps(data).encode('utf-8')

# Create a request object with the URL, data, and headers
request = urllib.request.Request(url, data=json_data, headers=headers, method='POST')

While this results in a 400 error

import urllib.request
import urllib.parse
import json

url = "http://localhost:11434/v1/completions"

headers = {
    'Content-Type': 'application/json'
}

data = {
    'prompt': "What is the cat holding?",
    'max_tokens': 20,
    'temperature': 1,
    'top_p': 0.9,
    'seed': 10
}

# Convert data to JSON format
json_data = json.dumps(data).encode('utf-8')

# Create a request object with the URL, data, and headers
request = urllib.request.Request(url, data=json_data, headers=headers, method='POST')

# Send the request and read the response
with urllib.request.urlopen(request) as response:
    response_data = response.read()

# If needed, decode the response data
response = json.loads(response_data.decode('utf-8'))

new_text = response["choices"][0]["text"]

print(new_text)

If possible, I would like for the second case to work, defaulting to one of the models ollama is currently running (Doesn't matter which).

Originally created by @balisujohn on GitHub (Jul 31, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/6089 This works: ```` url = "http://localhost:11434/v1/completions" headers = { 'Content-Type': 'application/json' } data = { 'model':"moondream", 'prompt': "What is the cat holding?", 'max_tokens': 20, 'temperature': 1, 'top_p': 0.9, 'seed': 10 } # Convert data to JSON format json_data = json.dumps(data).encode('utf-8') # Create a request object with the URL, data, and headers request = urllib.request.Request(url, data=json_data, headers=headers, method='POST') ```` While this results in a 400 error ```` import urllib.request import urllib.parse import json url = "http://localhost:11434/v1/completions" headers = { 'Content-Type': 'application/json' } data = { 'prompt': "What is the cat holding?", 'max_tokens': 20, 'temperature': 1, 'top_p': 0.9, 'seed': 10 } # Convert data to JSON format json_data = json.dumps(data).encode('utf-8') # Create a request object with the URL, data, and headers request = urllib.request.Request(url, data=json_data, headers=headers, method='POST') # Send the request and read the response with urllib.request.urlopen(request) as response: response_data = response.read() # If needed, decode the response data response = json.loads(response_data.decode('utf-8')) new_text = response["choices"][0]["text"] print(new_text) ```` If possible, I would like for the second case to work, defaulting to one of the models ollama is currently running (Doesn't matter which).
GiteaMirror added the feature request label 2026-05-03 22:52:08 -05:00
GiteaMirror commented 2026-05-03 22:52:13 -05:00
Author
Owner
Copy Link

@thinkverse commented on GitHub (Jul 31, 2024):

The model is required as per OpenAI API and since this is Ollama's compatibility API it needs to follow OpenAI's API reference. https://platform.openai.com/docs/api-reference/completions/create#completions-create-model

<!-- gh-comment-id:2260912531 --> @thinkverse commented on GitHub (Jul 31, 2024): The `model` is required as per OpenAI API and since this is Ollama's compatibility API it needs to follow OpenAI's API reference. https://platform.openai.com/docs/api-reference/completions/create#completions-create-model
GiteaMirror commented 2026-05-03 22:52:19 -05:00
Author
Owner
Copy Link

@royjhan commented on GitHub (Jul 31, 2024):

Indeed, model is a required field as specified by OpenAI documentation

<!-- gh-comment-id:2261029770 --> @royjhan commented on GitHub (Jul 31, 2024): Indeed, `model` is a required field as specified by OpenAI documentation
GiteaMirror commented 2026-05-03 22:52:25 -05:00
Author
Owner
Copy Link

@balisujohn commented on GitHub (Jul 31, 2024):

Ah interesting so it's actually text generation webui and koboldcpp being overly permissive.

<!-- gh-comment-id:2261037821 --> @balisujohn commented on GitHub (Jul 31, 2024): Ah interesting so it's actually text generation webui and koboldcpp being overly permissive.
GiteaMirror commented 2026-05-03 22:52:32 -05:00
Author
Owner
Copy Link

@royjhan commented on GitHub (Jul 31, 2024):

Yeah, I took a quick look on what they're doing and I see your need, but in providing compatibility endpoints we are focused on aligning as much as possible. A default model would need to be configured on the client-side.

<!-- gh-comment-id:2261039577 --> @royjhan commented on GitHub (Jul 31, 2024): Yeah, I took a quick look on what they're doing and I see your need, but in providing compatibility endpoints we are focused on aligning as much as possible. A default model would need to be configured on the client-side.
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 feature request
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#65843
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?