github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-07 00:22:43 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #14202] "error":"Invalid role: thinking" when using cloud gemini-3-pro-preview:latest #9254

New Issue
Open
opened 2026-04-12 22:07:51 -05:00 by GiteaMirror · 0 comments
GiteaMirror commented 2026-04-12 22:07:51 -05:00
Owner
Copy Link

Originally created by @eticolat on GitHub (Feb 11, 2026).
Original GitHub issue: https://github.com/ollama/ollama/issues/14202

What is the issue?

using continue dev extension in vscode, in chat or plan mode:

  1. submit your first prompt complex enough so the model answers with a think section
  2. submit your second prompt, the first prompt and answer (including the thinking) are included in the second prompt as context. A "thinking" role (in addition to system, user, assistant) is automatically added to the prompt with the first answer thoughts from the model.

The error occurs: Error HTTP 400 Bad Request from http://{ollama server url}/api/chat {"StatusCode":400,"Status":"400 Bad Request","error":"Invalid role: thinking"}

Same process using kimi-k2.5:cloud for the second prompt (exact same context including the thinking role): the request is properly answered by the model. No error.

Relevant log output

system
   system prompt ...
thinking
   thoughts from first model answer ...

assistant
   Hello I am ready to assist you ...
user
   second user prompt

ErrorHTTP 400 Bad Request from http://ollamaServerURL/api/chat {"StatusCode":400,"Status":"400 Bad Request","error":"Invalid role: thinking"}

OS

Docker

GPU

No response

CPU

No response

Ollama version

0.15.6

Originally created by @eticolat on GitHub (Feb 11, 2026). Original GitHub issue: https://github.com/ollama/ollama/issues/14202 ### What is the issue? using continue dev extension in vscode, in chat or plan mode: 1. submit your first prompt complex enough so the model answers with a think section 2. submit your second prompt, the first prompt and answer (including the thinking) are included in the second prompt as context. A "thinking" role (in addition to system, user, assistant) is automatically added to the prompt with the first answer thoughts from the model. The error occurs: Error HTTP 400 Bad Request from http://{ollama server url}/api/chat {"StatusCode":400,"Status":"400 Bad Request","error":"Invalid role: thinking"} Same process using kimi-k2.5:cloud for the second prompt (exact same context including the thinking role): the request is properly answered by the model. No error. ### Relevant log output ```shell system system prompt ... thinking thoughts from first model answer ... assistant Hello I am ready to assist you ... user second user prompt ErrorHTTP 400 Bad Request from http://ollamaServerURL/api/chat {"StatusCode":400,"Status":"400 Bad Request","error":"Invalid role: thinking"} ``` ### OS Docker ### GPU _No response_ ### CPU _No response_ ### Ollama version 0.15.6
GiteaMirror added the bug label 2026-04-12 22:07:51 -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 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#9254
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?