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 #9282] [CLOSED] runner: enable returning more info from runner processing #23450

New Issue
Closed
opened 2026-04-19 17:00:31 -05:00 by GiteaMirror · 0 comments
GiteaMirror commented 2026-04-19 17:00:31 -05:00
Owner
Copy Link

📋 Pull Request Information

Original PR: https://github.com/ollama/ollama/pull/9282
Author: @BruceMacD
Created: 2/22/2025
Status: ❌ Closed

Base: main ← Head: brucemacd/runner-completion

📝 Commits (1)

  • d5eae82 runner: enable returning more info from runner processing

📊 Changes

6 files changed (+376 additions, -228 deletions)

View changed files

📝 llm/server.go (+61 -2)
📝 llm/server_test.go (+149 -0)
📝 runner/common/stop.go (+27 -52)
📝 runner/common/stop_test.go (+68 -88)
📝 runner/llamarunner/runner.go (+35 -43)
📝 runner/ollamarunner/runner.go (+36 -43)

📄 Description

This refactor enables returning additional information from the LLM runner beyond just text. Currently, we only return the raw prediction text, but we need more flexibility to include metadata like logprobs.

The change introduces moves the CompletionResponse struct to the common package and converts the internal response handling from simple strings to these structured objects. All the runner code (both llama and ollama implementations) now works with this structure while maintaining the same text processing logic.

With this foundation in place, my follow-up PR can easily add logprobs to the response during batch processing, when tokens are actually being generated, without needing to restructure the whole pipeline again.

🔄 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/9282 **Author:** [@BruceMacD](https://github.com/BruceMacD) **Created:** 2/22/2025 **Status:** ❌ Closed **Base:** `main` ← **Head:** `brucemacd/runner-completion` --- ### 📝 Commits (1) - [`d5eae82`](https://github.com/ollama/ollama/commit/d5eae8248dea951b82dca0a9d72b4800f629700f) runner: enable returning more info from runner processing ### 📊 Changes **6 files changed** (+376 additions, -228 deletions) <details> <summary>View changed files</summary> 📝 `llm/server.go` (+61 -2) 📝 `llm/server_test.go` (+149 -0) 📝 `runner/common/stop.go` (+27 -52) 📝 `runner/common/stop_test.go` (+68 -88) 📝 `runner/llamarunner/runner.go` (+35 -43) 📝 `runner/ollamarunner/runner.go` (+36 -43) </details> ### 📄 Description This refactor enables returning additional information from the LLM runner beyond just text. Currently, we only return the raw prediction text, but we need more flexibility to include metadata like logprobs. The change introduces moves the CompletionResponse struct to the common package and converts the internal response handling from simple strings to these structured objects. All the runner code (both llama and ollama implementations) now works with this structure while maintaining the same text processing logic. With this foundation in place, my follow-up PR can easily add logprobs to the response during batch processing, when tokens are actually being generated, without needing to restructure the whole pipeline again. --- <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 17:00:31 -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#23450
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?