github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-07 16:40:08 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #7910] tool parsing issues with "'" #5062

New Issue
Open
opened 2026-04-12 16:09:13 -05:00 by GiteaMirror · 5 comments
GiteaMirror commented 2026-04-12 16:09:13 -05:00
Owner
Copy Link

Originally created by @fce2 on GitHub (Dec 2, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/7910

What is the issue?

difficult to see in the title: ' is the problem.
when i ask my ai to "execute a python example" it generates something like "print('...')" but truncates at the 1st ':

    "model": "llama3.1:8b-instruct-fp16",
    "created_at": "2024-12-02T13:26:55.1045197Z",
    "message": {
        "role": "assistant",
        "content": "",
        "tool_calls": [
            {
                "function": {
                    "name": "execute_python",
                    "arguments": {
                        "code": "print("
                    }
                }
            }
        ]
    },
    "done_reason": "stop",
    "done": true,
    "total_duration": 574508000,
    "load_duration": 10134500,
    "prompt_eval_count": 1371,
    "prompt_eval_duration": 3000000,
    "eval_count": 18,
    "eval_duration": 559000000

OS

No response

GPU

No response

CPU

No response

Ollama version

No response

Originally created by @fce2 on GitHub (Dec 2, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/7910 ### What is the issue? difficult to see in the title: ' is the problem. when i ask my ai to "execute a python example" it generates something like "print('...')" but truncates at the 1st ': "model": "llama3.1:8b-instruct-fp16", "created_at": "2024-12-02T13:26:55.1045197Z", "message": { "role": "assistant", "content": "", "tool_calls": [ { "function": { "name": "execute_python", "arguments": { "code": "print(" } } } ] }, "done_reason": "stop", "done": true, "total_duration": 574508000, "load_duration": 10134500, "prompt_eval_count": 1371, "prompt_eval_duration": 3000000, "eval_count": 18, "eval_duration": 559000000 ### OS _No response_ ### GPU _No response_ ### CPU _No response_ ### Ollama version _No response_
GiteaMirror added the bug label 2026-04-12 16:09:13 -05:00
GiteaMirror commented 2026-04-12 16:09:14 -05:00
Author
Owner
Copy Link

@fce2 commented on GitHub (Dec 2, 2024):

maybe its not the ',my js executor works:

    "model": "llama3.1:8b-instruct-fp16",
    "created_at": "2024-12-02T13:32:11.9342824Z",
    "message": {
        "role": "assistant",
        "content": "",
        "tool_calls": [
            {
                "function": {
                    "name": "execute_javascript",
                    "arguments": {
                        "code": "console.log('Hello World!');"
                    }
                }
            }
        ]
    },
    "done_reason": "stop",
    "done": true,
    "total_duration": 720423100,
    "load_duration": 11974700,
    "prompt_eval_count": 1371,
    "prompt_eval_duration": 242000000,
    "eval_count": 24,
    "eval_duration": 464000000
<!-- gh-comment-id:2511558991 --> @fce2 commented on GitHub (Dec 2, 2024): maybe its not the ',my js executor works: "model": "llama3.1:8b-instruct-fp16", "created_at": "2024-12-02T13:32:11.9342824Z", "message": { "role": "assistant", "content": "", "tool_calls": [ { "function": { "name": "execute_javascript", "arguments": { "code": "console.log('Hello World!');" } } } ] }, "done_reason": "stop", "done": true, "total_duration": 720423100, "load_duration": 11974700, "prompt_eval_count": 1371, "prompt_eval_duration": 242000000, "eval_count": 24, "eval_duration": 464000000
GiteaMirror commented 2026-04-12 16:09:14 -05:00
Author
Owner
Copy Link

@fce2 commented on GitHub (Dec 2, 2024):

sorry, looks like its no issue with ollama but with the model i use.
if i rephrade my query it works:

        "role": "assistant",
        "content": "",
        "tool_calls": [
            {
                "function": {
                    "name": "execute_python",
                    "arguments": {
                        "code": "print('Hello World!')"
                    }
                }
            }
        ]
    },
<!-- gh-comment-id:2511571280 --> @fce2 commented on GitHub (Dec 2, 2024): sorry, looks like its no issue with ollama but with the model i use. if i rephrade my query it works: "role": "assistant", "content": "", "tool_calls": [ { "function": { "name": "execute_python", "arguments": { "code": "print('Hello World!')" } } } ] },
GiteaMirror commented 2026-04-12 16:09:15 -05:00
Author
Owner
Copy Link

@fce2 commented on GitHub (Dec 2, 2024):

closed

<!-- gh-comment-id:2511618554 --> @fce2 commented on GitHub (Dec 2, 2024): closed
GiteaMirror commented 2026-04-12 16:09:15 -05:00
Author
Owner
Copy Link

@fce2 commented on GitHub (Dec 9, 2024):

I have to reopen this issue.
Sometimes (?) the anser is cut when a ' should come

here again:
"model": "llama3.1:8b-instruct-q8_0-32768",
"created_at": "2024-12-09T21:17:54.0578108Z",
"message": {
"role": "assistant",
"content": "",
"tool_calls": [
{
"function": {
"name": "direct_answer",
"arguments": {
"answer": "I don"
}
}
}
]
},

<!-- gh-comment-id:2529506529 --> @fce2 commented on GitHub (Dec 9, 2024): I have to reopen this issue. Sometimes (?) the anser is cut when a ' should come here again: "model": "llama3.1:8b-instruct-q8_0-32768", "created_at": "2024-12-09T21:17:54.0578108Z", "message": { "role": "assistant", "content": "", "tool_calls": [ { "function": { "name": "direct_answer", "arguments": { "answer": "I don" } } } ] },
GiteaMirror commented 2026-04-12 16:09:16 -05:00
Author
Owner
Copy Link

@scionaltera commented on GitHub (Apr 19, 2025):

I'm seeing this too, running latest ollama in Docker on Windows 11 with llama 3.2. Every time the model's response to the tool call would contain an apostrophe the response is cut off there. Seems like a JSON parsing issue but I'm not anywhere close to familiar enough with the ollama code yet to identify where. It only happens with tool calls, not regular chat messages.

As an example, I have a tool for storing a "memory" to a database. If the model wanted to store something like "Remember to take Ben's bike to Cub Scouts next Wednesday." the result would just be "Remember to take Ben".

<!-- gh-comment-id:2816905388 --> @scionaltera commented on GitHub (Apr 19, 2025): I'm seeing this too, running latest ollama in Docker on Windows 11 with llama 3.2. Every time the model's response to the tool call would contain an apostrophe the response is cut off there. Seems like a JSON parsing issue but I'm not anywhere close to familiar enough with the ollama code yet to identify where. It only happens with tool calls, not regular chat messages. As an example, I have a tool for storing a "memory" to a database. If the model wanted to store something like "Remember to take Ben's bike to Cub Scouts next Wednesday." the result would just be "Remember to take Ben".
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#5062
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?