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 #13581] Feature Request: Stream tool call arguments progressively #8942

New Issue
Open
opened 2026-04-12 21:45:51 -05:00 by GiteaMirror · 1 comment
GiteaMirror commented 2026-04-12 21:45:51 -05:00
Owner
Copy Link

Originally created by @Baehn on GitHub (Dec 29, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/13581

In issue #11633, it was confirmed that Ollama currently sends each tool call as a single, fully assembled JSON chunk.

This behavior becomes problematic for tools where argument generation may take a long time — for example, a create_document tool where the full document content is included in the arguments. In such cases, downstream clients and UIs receive no incremental updates and must wait until the entire argument payload is completed.

It would be great if Ollama could support progressive streaming of tool call arguments, rather than sending the entire tool call as one complete chunk. This would work like message streaming does now, letting clients build up long tool inputs piece by piece and show progress as it comes in. The behavior should be opt‑in, for example through a parameter like stream_tool_calls=True, so existing integrations remain unaffected.

Originally created by @Baehn on GitHub (Dec 29, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/13581 In issue #11633, it was confirmed that Ollama currently sends each tool call as a single, fully assembled JSON chunk. This behavior becomes problematic for tools where argument generation may take a long time — for example, a `create_document` tool where the full document content is included in the arguments. In such cases, downstream clients and UIs receive no incremental updates and must wait until the entire argument payload is completed. It would be great if Ollama could support progressive streaming of tool call arguments, rather than sending the entire tool call as one complete chunk. This would work like message streaming does now, letting clients build up long tool inputs piece by piece and show progress as it comes in. The behavior should be opt‑in, for example through a parameter like `stream_tool_calls=True`, so existing integrations remain unaffected.
GiteaMirror added the feature request label 2026-04-12 21:45:51 -05:00
GiteaMirror commented 2026-04-12 21:45:53 -05:00
Author
Owner
Copy Link

@dragocz1995 commented on GitHub (Mar 26, 2026):

Same problem, all other providers mostly have it. Thank you!

<!-- gh-comment-id:4134779084 --> @dragocz1995 commented on GitHub (Mar 26, 2026): Same problem, all other providers mostly have it. Thank you!
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#8942
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?