github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-07 08:30:05 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[PR #2410] [CLOSED] Added Encoding endpoint #21420

New Issue
Closed
opened 2026-04-19 15:37:24 -05:00 by GiteaMirror · 0 comments
GiteaMirror commented 2026-04-19 15:37:24 -05:00
Owner
Copy Link

📋 Pull Request Information

Original PR: https://github.com/ollama/ollama/pull/2410
Author: @suvalaki
Created: 2/8/2024
Status: ❌ Closed

Base: main ← Head: main

📝 Commits (1)

📊 Changes

3 files changed (+169 additions, -0 deletions)

View changed files

📝 api/client.go (+13 -0)
📝 api/types.go (+10 -0)
📝 server/routes.go (+146 -0)

📄 Description

It seems useful to expose the encoding function of a model that us called by the generate methods to enable token counting (without running the model end to end).

Some thoughts:

  • Im not sure whether replicating the logic that modifies the prompt (the same as the generate function is correct here or whether we should just simplify and just look at the raw prompt string.
  • I havent made a similar endpoint for getting the tokens from a chat call.

My current thinking is that simplification of this would be better and i dont need replicate all the prompt logic necessarily. Im interested in feedback and if folks have any pushback to exposing the function.

Following up from https://github.com/ollama/ollama/issues/1345

Btw the proposal makes a new endpoint /api/encode for requests of the following

Looks a bit like this at a request level:
Input

{
  "model": "mistral:latest",
  "prompt": "Why is the sky blue?"
}

Output

{
    "model": "mistral:latest",
    "created_at": "2024-02-05T21:49:44.472893Z",
    "total_duration": 8965307875,
    "load_duration": 8961889917,
    "context": [
        733,
        16289,
        28793,
        28705,
        4315,
        349,
        272,
        7212,
        5045,
        28804,
        733,
        28748,
        16289,
        28793,
        13
    ],
    "prompt_eval_count": 15
}

🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.

## 📋 Pull Request Information **Original PR:** https://github.com/ollama/ollama/pull/2410 **Author:** [@suvalaki](https://github.com/suvalaki) **Created:** 2/8/2024 **Status:** ❌ Closed **Base:** `main` ← **Head:** `main` --- ### 📝 Commits (1) - [`60d90ca`](https://github.com/ollama/ollama/commit/60d90ca094c68ef33c93e904eb412d06d5f6bffc) Added Encoding endpoint ### 📊 Changes **3 files changed** (+169 additions, -0 deletions) <details> <summary>View changed files</summary> 📝 `api/client.go` (+13 -0) 📝 `api/types.go` (+10 -0) 📝 `server/routes.go` (+146 -0) </details> ### 📄 Description It seems useful to expose the encoding function of a model that us called by the generate methods to enable token counting (without running the model end to end). Some thoughts: - Im not sure whether replicating the logic that modifies the prompt (the same as the generate function is correct here or whether we should just simplify and just look at the raw prompt string. - I havent made a similar endpoint for getting the tokens from a chat call. My current thinking is that simplification of this would be better and i dont need replicate all the prompt logic necessarily. Im interested in feedback and if folks have any pushback to exposing the function. Following up from https://github.com/ollama/ollama/issues/1345 Btw the proposal makes a new endpoint `/api/encode` for requests of the following Looks a bit like this at a request level: Input ```json { "model": "mistral:latest", "prompt": "Why is the sky blue?" } ``` Output ```json { "model": "mistral:latest", "created_at": "2024-02-05T21:49:44.472893Z", "total_duration": 8965307875, "load_duration": 8961889917, "context": [ 733, 16289, 28793, 28705, 4315, 349, 272, 7212, 5045, 28804, 733, 28748, 16289, 28793, 13 ], "prompt_eval_count": 15 } ``` --- <sub>🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.</sub>
GiteaMirror added the pull-request label 2026-04-19 15:37:24 -05:00
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 pull-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#21420
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?