github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-08 17:49:24 -05:00
Code Issues 14.6k Packages Projects Releases 122 Wiki Activity

[GH-ISSUE #13755] CLI multiline prompt shows no response until follow-up message is sent #71076

New Issue
Closed
opened 2026-05-04 23:55:45 -05:00 by GiteaMirror · 1 comment
GiteaMirror commented 2026-05-04 23:55:45 -05:00
Owner
Copy Link

Originally created by @timlucastech on GitHub (Jan 17, 2026).
Original GitHub issue: https://github.com/ollama/ollama/issues/13755

What is the issue?

When entering a multiline prompt in interactive mode (using the ... continuation lines), the model processes the request but no response is displayed. The response only appears after sending a follow-up message (e.g., "Hello?"), at which point the model's response references the original prompt, proving it was processed.

OS

Linux (Ubuntu)

GPU

CPU-only (also has AMD GPU with ROCm installed)

CPU

x86_64

Ollama version

0.14.2

Model

danielsheep/Qwen3-Coder-30B-A3B-Instruct-1M-Unsloth:UD-Q5_K_XL

Steps to reproduce

  1. Run ollama run (tested with danielsheep/Qwen3-Coder-30B-A3B-Instruct-1M-Unsloth:UD-Q5_K_XL)
  2. Enter a multiline prompt that spans multiple lines:

Generate a comprehensive inline HTML interface that uses tailwind CDNs and fetches the latest price of bitcoin and prints it to the interface; make the interface nice and modern and smoothly update in realtime by refreshing the GUI without flickers or visual glitches; make
... the interface modern and aesthetically nice UX

  1. Press Enter - cursor returns to >>> prompt with no response displayed
  2. Type any follow-up message like "Hello?"
  3. The model now responds, referencing the original request (e.g., "I can see you're interested in the Bitcoin price tracker HTML file I generated for you...")

Expected behavior

Response should stream/display immediately after the multiline prompt is submitted.

Actual behavior

No response is displayed. The response only appears after a subsequent message is sent.

Additional context

  • Single-line prompts in interactive mode work correctly
  • Piped input works correctly: echo "prompt" | ollama run
  • API calls work correctly: curl http://localhost:11434/api/generate -d '...'
  • This suggests the issue is specific to the readline/terminal handling for multiline interactive input
  • Terminal: xterm-256color
  • Issue persists after upgrading from 0.14.1 to 0.14.2

Logs

Server logs show successful HTTP 200 responses for the requests, confirming the model processed them:
[GIN] 2026/01/17 - 14:09:42 | 200 | 796.41043ms | 127.0.0.1 | POST "/api/chat"
[GIN] 2026/01/17 - 14:10:54 | 200 | 5.754584434s | 127.0.0.1 | POST "/api/chat"

Originally created by @timlucastech on GitHub (Jan 17, 2026). Original GitHub issue: https://github.com/ollama/ollama/issues/13755 What is the issue? When entering a multiline prompt in interactive mode (using the ... continuation lines), the model processes the request but no response is displayed. The response only appears after sending a follow-up message (e.g., "Hello?"), at which point the model's response references the original prompt, proving it was processed. OS Linux (Ubuntu) GPU CPU-only (also has AMD GPU with ROCm installed) CPU x86_64 Ollama version 0.14.2 Model danielsheep/Qwen3-Coder-30B-A3B-Instruct-1M-Unsloth:UD-Q5_K_XL Steps to reproduce 1. Run ollama run <model> (tested with danielsheep/Qwen3-Coder-30B-A3B-Instruct-1M-Unsloth:UD-Q5_K_XL) 2. Enter a multiline prompt that spans multiple lines: >>> Generate a comprehensive inline HTML interface that uses tailwind CDNs and fetches the latest price of bitcoin and prints it to the interface; make the interface nice and modern and smoothly update in realtime by refreshing the GUI without flickers or visual glitches; make ... the interface modern and aesthetically nice UX 3. Press Enter - cursor returns to >>> prompt with no response displayed 4. Type any follow-up message like "Hello?" 5. The model now responds, referencing the original request (e.g., "I can see you're interested in the Bitcoin price tracker HTML file I generated for you...") Expected behavior Response should stream/display immediately after the multiline prompt is submitted. Actual behavior No response is displayed. The response only appears after a subsequent message is sent. Additional context - Single-line prompts in interactive mode work correctly - Piped input works correctly: echo "prompt" | ollama run <model> - API calls work correctly: curl http://localhost:11434/api/generate -d '...' - This suggests the issue is specific to the readline/terminal handling for multiline interactive input - Terminal: xterm-256color - Issue persists after upgrading from 0.14.1 to 0.14.2 Logs Server logs show successful HTTP 200 responses for the requests, confirming the model processed them: [GIN] 2026/01/17 - 14:09:42 | 200 | 796.41043ms | 127.0.0.1 | POST "/api/chat" [GIN] 2026/01/17 - 14:10:54 | 200 | 5.754584434s | 127.0.0.1 | POST "/api/chat"
GiteaMirror added the bugneeds more info labels 2026-05-04 23:55:45 -05:00
GiteaMirror commented 2026-05-04 23:55:54 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Jan 17, 2026):

Unable to reproduce. The short response time (796ms) implies that the model received the prompt but ignored it. On an AMD Ryzen 9 7900X system the model took 88s to generate the response. Try setting OLLAMA_DEBUG=2 and post the log, this will show the prompt and might provide insight in to what's going on.

<!-- gh-comment-id:3762964397 --> @rick-github commented on GitHub (Jan 17, 2026): Unable to reproduce. The short response time (796ms) implies that the model received the prompt but ignored it. On an AMD Ryzen 9 7900X system the model took 88s to generate the response. Try setting `OLLAMA_DEBUG=2` and post the log, this will show the prompt and might provide insight in to what's going on.
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 needs more info
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#71076
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?