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 #10015] Final stream chunk missing "done": true despite completed response #6567

New Issue
Closed
opened 2026-04-12 18:11:44 -05:00 by GiteaMirror · 2 comments
GiteaMirror commented 2026-04-12 18:11:44 -05:00
Owner
Copy Link

Originally created by @jopersr on GitHub (Mar 27, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/10015

What is the issue?

In certain cases when using the /generate endpoint with "stream": true, the final chunk of the response stream includes all expected metadata (like infoContext and chatId), but the "done": true flag is missing — leaving the client unsure whether the stream has truly ended.

Relevant log output

curl --request POST \
  --url http://localhost:11434/api/generate \
  --header 'Content-Type: application/json' \
  --data '{
    "title": "Title028",
    "options": {},
    "prompt": "Cuentame una historia de 50 parrafos sobre un dragon y un a princesa. En la histyoria debe aparecer un unicornio, un caballero y una matiposa magica. Esta historia sr[a para trabajar literatura con mis estudiantes de 8 años.",
    "stream": true
}'

Expected output (trimmed):
{ "response": "Once", "done": false }
{ "response": " upon", "done": false }
...
{ "response": ".", "done": true, "infoContext": {...}, "chatId": "..." }

Actual output (problematic case):
{ "response": ".", "done": false }
{ "response": ".", "done": false }
{ "response": ".", "done": false, "infoContext": {...}, "chatId": "..." }

Clients consuming the stream rely on the done: true flag to trigger proper UI updates and finalize processing.
When it is omitted, the frontend is left in an uncertain state, waiting for more chunks that will never come.

Even though the stream ends and metadata is received, the client is never officially told, “You’re done.”

OS

Linux

GPU

Nvidia

CPU

No response

Ollama version

llama3.1:8b

Originally created by @jopersr on GitHub (Mar 27, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/10015 ### What is the issue? In certain cases when using the /generate endpoint with "stream": true, the final chunk of the response stream includes all expected metadata (like infoContext and chatId), but the "done": true flag is missing — leaving the client unsure whether the stream has truly ended. ### Relevant log output ```shell curl --request POST \ --url http://localhost:11434/api/generate \ --header 'Content-Type: application/json' \ --data '{ "title": "Title028", "options": {}, "prompt": "Cuentame una historia de 50 parrafos sobre un dragon y un a princesa. En la histyoria debe aparecer un unicornio, un caballero y una matiposa magica. Esta historia sr[a para trabajar literatura con mis estudiantes de 8 años.", "stream": true }' Expected output (trimmed): { "response": "Once", "done": false } { "response": " upon", "done": false } ... { "response": ".", "done": true, "infoContext": {...}, "chatId": "..." } Actual output (problematic case): { "response": ".", "done": false } { "response": ".", "done": false } { "response": ".", "done": false, "infoContext": {...}, "chatId": "..." } Clients consuming the stream rely on the done: true flag to trigger proper UI updates and finalize processing. When it is omitted, the frontend is left in an uncertain state, waiting for more chunks that will never come. Even though the stream ends and metadata is received, the client is never officially told, “You’re done.” ``` ### OS Linux ### GPU Nvidia ### CPU _No response_ ### Ollama version llama3.1:8b
GiteaMirror added the bug label 2026-04-12 18:11:44 -05:00
GiteaMirror commented 2026-04-12 18:11:45 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Mar 27, 2025):

Actual output (problematic case):
{ "response": ".", "done": false }
{ "response": ".", "done": false }
{ "response": ".", "done": false, "infoContext": {...}, "chatId": "..." }

This is not output from ollama - there are no "infoContext" or "chatId" fields in ollama responses.

<!-- gh-comment-id:2758647441 --> @rick-github commented on GitHub (Mar 27, 2025): ``` Actual output (problematic case): { "response": ".", "done": false } { "response": ".", "done": false } { "response": ".", "done": false, "infoContext": {...}, "chatId": "..." } ``` This is not output from ollama - there are no "infoContext" or "chatId" fields in ollama responses.
GiteaMirror commented 2026-04-12 18:11:46 -05:00
Author
Owner
Copy Link

@jopersr commented on GitHub (Mar 27, 2025):

Was my fault, the problem was in our backend . Thanks for your response.

<!-- gh-comment-id:2759054745 --> @jopersr commented on GitHub (Mar 27, 2025): Was my fault, the problem was in our backend . Thanks for your response.
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 bug
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#6567
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?