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 #687] Error: error reading llm response: bufio.Scanner: token too long #26073

New Issue
Closed
opened 2026-04-22 01:58:47 -05:00 by GiteaMirror · 4 comments
GiteaMirror commented 2026-04-22 01:58:47 -05:00
Owner
Copy Link

Originally created by @reggi on GitHub (Oct 3, 2023).
Original GitHub issue: https://github.com/ollama/ollama/issues/687

Originally assigned to: @BruceMacD on GitHub.

I'm passing in about 62 articles that I wrote from the web and trying to get some analysis on them, and I keep seeing this error:

Error: error reading llm response: bufio.Scanner: token too long

Some text response comes back and then after a couple sentences it throws that error.

Is there a better way to do this or is my machine just not capable? Running a M1 Macbook air.

function generate_output() {
  echo -e "Below is a series of articles that I wrote\n"
  echo -e "\nEach article is prefixed with the string 'Here's another article:'\n"
  echo -e "\nStarting Articles\n"

  files=(*.txt)
  count=${#files[@]}
  i=0

  for file in "${files[@]}"; do
    cat "$file"
    ((i++))
    if [ $i -lt $count ]; then
      echo -e "\nHere's another article:\n"
    fi
  done

  echo -e "\nEnd of Articles\n"
  echo -e "\nCan you interpret some common themes from this series of articles?\n"
  # echo -e "\nWhat is the core theme from these articles?\n"
  # echo -e "\nWrite a bio for this author.\n"
}

ollama run mistral "$(generate_output)"
# ollama run codellama "$(generate_output)"
Originally created by @reggi on GitHub (Oct 3, 2023). Original GitHub issue: https://github.com/ollama/ollama/issues/687 Originally assigned to: @BruceMacD on GitHub. I'm passing in about 62 articles that I wrote from the web and trying to get some analysis on them, and I keep seeing this error: > `Error: error reading llm response: bufio.Scanner: token too long` Some text response comes back and then after a couple sentences it throws that error. Is there a better way to do this or is my machine just not capable? Running a M1 Macbook air. ```bash function generate_output() { echo -e "Below is a series of articles that I wrote\n" echo -e "\nEach article is prefixed with the string 'Here's another article:'\n" echo -e "\nStarting Articles\n" files=(*.txt) count=${#files[@]} i=0 for file in "${files[@]}"; do cat "$file" ((i++)) if [ $i -lt $count ]; then echo -e "\nHere's another article:\n" fi done echo -e "\nEnd of Articles\n" echo -e "\nCan you interpret some common themes from this series of articles?\n" # echo -e "\nWhat is the core theme from these articles?\n" # echo -e "\nWrite a bio for this author.\n" } ollama run mistral "$(generate_output)" # ollama run codellama "$(generate_output)" ```
GiteaMirror commented 2026-04-22 01:58:49 -05:00
Author
Owner
Copy Link

@BruceMacD commented on GitHub (Oct 3, 2023):

Thanks for reporting this. I've opened #692 to get the error fixed.

As a side note, the context window for the LLM wont be able to fit a lot of long articles, they will be automatically truncated. A better approach to this would be breaking this down into individual requests, then sending a final request that asks the LLM to create a summary based on those shorter outputs.

<!-- gh-comment-id:1745678880 --> @BruceMacD commented on GitHub (Oct 3, 2023): Thanks for reporting this. I've opened #692 to get the error fixed. As a side note, the context window for the LLM wont be able to fit a lot of long articles, they will be automatically truncated. A better approach to this would be breaking this down into individual requests, then sending a final request that asks the LLM to create a summary based on those shorter outputs.
GiteaMirror commented 2026-04-22 01:58:49 -05:00
Author
Owner
Copy Link

@reggi commented on GitHub (Oct 3, 2023):

@BruceMacD thanks for the feedback, just to confirm I should do this via curl and not piping / arguments via shell script? memory is able to persist across curl requests?

<!-- gh-comment-id:1745825370 --> @reggi commented on GitHub (Oct 3, 2023): @BruceMacD thanks for the feedback, just to confirm I should do this via curl and not piping / arguments via shell script? memory is able to persist across curl requests?
GiteaMirror commented 2026-04-22 01:58:50 -05:00
Author
Owner
Copy Link

@BruceMacD commented on GitHub (Oct 4, 2023):

You could still use piping. I'd just suggest breaking up the piped requests into smaller prompts so that the model doesn't lose some context.

<!-- gh-comment-id:1747312151 --> @BruceMacD commented on GitHub (Oct 4, 2023): You could still use piping. I'd just suggest breaking up the piped requests into smaller prompts so that the model doesn't lose some context.
GiteaMirror commented 2026-04-22 01:58:51 -05:00
Author
Owner
Copy Link

@reggi commented on GitHub (Oct 4, 2023):

@BruceMacD from expierence new lines \n are split and each line is sent as a prompt I had no way of providing blocks of text with new lines as a single prompt via piping.

<!-- gh-comment-id:1747474821 --> @reggi commented on GitHub (Oct 4, 2023): @BruceMacD from expierence new lines \n are split and each line is sent as a prompt I had no way of providing blocks of text with new lines as a single prompt via piping.
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
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#26073
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?