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 #6592] Model whitelisting for generate endpoint #4150

New Issue
Open
opened 2026-04-12 15:03:47 -05:00 by GiteaMirror · 2 comments
GiteaMirror commented 2026-04-12 15:03:47 -05:00
Owner
Copy Link

Originally created by @JTHesse on GitHub (Sep 2, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/6592

I would like to support the ability to whitelist models that are usable with the generate endpoint.

The model list should be provided via an env variable and be checked against before running the generate function.

With this feature, the admins will gain more control over the models that can used by auto coding, f.e.: continue or other services. High frequent request from those applications will lead to huge delays, if the wrong model is selected.

Originally created by @JTHesse on GitHub (Sep 2, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/6592 I would like to support the ability to whitelist models that are usable with the generate endpoint. The model list should be provided via an env variable and be checked against before running the generate function. With this feature, the admins will gain more control over the models that can used by auto coding, f.e.: [continue ](https://www.continue.dev/) or other services. High frequent request from those applications will lead to huge delays, if the wrong model is selected.
GiteaMirror added the feature request label 2026-04-12 15:03:47 -05:00
GiteaMirror commented 2026-04-12 15:03:48 -05:00
Author
Owner
Copy Link

@adamoutler commented on GitHub (Mar 25, 2025):

Looking at this code https://github.com/ollama/ollama/blob/main/cmd/cmd.go#L775

I think it could have an environmental variable added

...
    environment:
      OLLAMA_ALLOWED_MODEL_PREFIX: gemma,granite
...

By implementing like this

package main

import (
    "net/http"
    "os"
    "strings"
)

// PullHandler handles requests to pull models
func PullHandler(w http.ResponseWriter, r *http.Request) {
    // Retrieve the allowed model prefixes from the environment variable
    allowedPrefixes := strings.Split(os.Getenv("OLLAMA_ALLOWED_MODEL_PREFIX"), ",")

    // Extract the model name from the request parameters
    modelName := r.URL.Query().Get("model")
    if modelName == "" {
        http.Error(w, "Model name is required", http.StatusBadRequest)
        return
    }

    // Check if the model name starts with any of the allowed prefixes
    isAllowed := false
    for _, prefix := range allowedPrefixes {
        if strings.HasPrefix(modelName, prefix) {
            isAllowed = true
            break
        }
    }

    if !isAllowed {
        http.Error(w, "Model not allowed", http.StatusForbidden)
        return
    }

    // Proceed with the existing pull logic for allowed models
    // ...
}
<!-- gh-comment-id:2750943355 --> @adamoutler commented on GitHub (Mar 25, 2025): Looking at this code https://github.com/ollama/ollama/blob/main/cmd/cmd.go#L775 I think it could have an environmental variable added ``` docker-compose ... environment: OLLAMA_ALLOWED_MODEL_PREFIX: gemma,granite ... ``` By implementing like this ``` go package main import ( "net/http" "os" "strings" ) // PullHandler handles requests to pull models func PullHandler(w http.ResponseWriter, r *http.Request) { // Retrieve the allowed model prefixes from the environment variable allowedPrefixes := strings.Split(os.Getenv("OLLAMA_ALLOWED_MODEL_PREFIX"), ",") // Extract the model name from the request parameters modelName := r.URL.Query().Get("model") if modelName == "" { http.Error(w, "Model name is required", http.StatusBadRequest) return } // Check if the model name starts with any of the allowed prefixes isAllowed := false for _, prefix := range allowedPrefixes { if strings.HasPrefix(modelName, prefix) { isAllowed = true break } } if !isAllowed { http.Error(w, "Model not allowed", http.StatusForbidden) return } // Proceed with the existing pull logic for allowed models // ... } ```
GiteaMirror commented 2026-04-12 15:03:49 -05:00
Author
Owner
Copy Link

@adamoutler commented on GitHub (Mar 25, 2025):

I think this is an important feature for government/enterprise

<!-- gh-comment-id:2750946803 --> @adamoutler commented on GitHub (Mar 25, 2025): I think this is an important feature for government/enterprise
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#4150
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?