github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-06 08:02:14 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[PR #7453] [MERGED] runner.go: Don't set cross attention before sending embeddings #38299

New Issue
Closed
opened 2026-04-22 22:58:20 -05:00 by GiteaMirror · 0 comments
GiteaMirror commented 2026-04-22 22:58:20 -05:00
Owner
Copy Link

📋 Pull Request Information

Original PR: https://github.com/ollama/ollama/pull/7453
Author: @jessegross
Created: 10/31/2024
Status: ✅ Merged
Merged: 10/31/2024
Merged by: @jessegross

Base: main ← Head: jessegross/cross_attn

📝 Commits (1)

  • 42e5133 runner.go: Don't set cross attention before sending embeddings

📊 Changes

2 files changed (+23 additions, -9 deletions)

View changed files

📝 llama/runner/image.go (+11 -0)
📝 llama/runner/runner.go (+12 -9)

📄 Description

Currently if an input has embeddings at any point then we will set cross attention to true from the beginning. This means that any tokens before the embeddings are sent will incorrectly have cross attention layers applied.

This only sets cross attention when we have an embedding, either previously in this sequence or in the cache. It also makes cross attention capable of supporting parallelism at the runner level, though the mllama implementation doesn't support that yet.

🔄 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/7453 **Author:** [@jessegross](https://github.com/jessegross) **Created:** 10/31/2024 **Status:** ✅ Merged **Merged:** 10/31/2024 **Merged by:** [@jessegross](https://github.com/jessegross) **Base:** `main` ← **Head:** `jessegross/cross_attn` --- ### 📝 Commits (1) - [`42e5133`](https://github.com/ollama/ollama/commit/42e5133d9bac494ac8f62c18773719ed330805e4) runner.go: Don't set cross attention before sending embeddings ### 📊 Changes **2 files changed** (+23 additions, -9 deletions) <details> <summary>View changed files</summary> 📝 `llama/runner/image.go` (+11 -0) 📝 `llama/runner/runner.go` (+12 -9) </details> ### 📄 Description Currently if an input has embeddings at any point then we will set cross attention to true from the beginning. This means that any tokens before the embeddings are sent will incorrectly have cross attention layers applied. This only sets cross attention when we have an embedding, either previously in this sequence or in the cache. It also makes cross attention capable of supporting parallelism at the runner level, though the mllama implementation doesn't support that yet. --- <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 2026-04-22 22:58:20 -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#38299
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?