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 #12200] The data that should be in the tool_calls section comes in the content section #33875

New Issue
Closed
opened 2026-04-22 17:00:57 -05:00 by GiteaMirror · 5 comments
GiteaMirror commented 2026-04-22 17:00:57 -05:00
Owner
Copy Link

Originally created by @docjon09 on GitHub (Sep 6, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/12200

What is the issue?

When I run the api request with the following body

{
  "model": "hermes-3-8b:latest",
  "messages": [
    {
      "role": "user",
      "content": "List all books"
    }
  ],
  "tools": [
    {
        "type": "function",
        "function": {
            "name": "fetch_product_by_entity_code_and_entity_key_code",
            "description": "Fetch the product with given entity code and given entity key code",
            "parameters": {
                "type": "object",
                "properties": {
                    "entityCode": {
                        "type": "string",
                        "description": "The entity code of the product"
                    },
                    "entityKeyCode": {
                        "type": "string",
                        "description": "The entity key code of the product"
                    }
                },
                "required": ["entityCode", "entityKeyCode"]
            }
        }
    },
    {
        "type": "function",
        "function": {
            "name": "fetch_all_books",
            "description": "Fetch all the books",
            "parameters": {}
        }
    }
  ],
  "options": {
    "temperature": 0
  },
  "stream": false,
}

The output is as follows

{
    "model": "hermes-3-8b:latest",
    "created_at": "2025-09-06T10:53:48.80640778Z",
    "message": {
        "role": "assistant",
        "content": "[\n  {\"name\":\"fetch_all_books\",\"arguments\":{}}\n]"
    },
    "done_reason": "stop",
    "done": true,
    "total_duration": 8895005988,
    "load_duration": 70760242,
    "prompt_eval_count": 266,
    "prompt_eval_duration": 7481875206,
    "eval_count": 15,
    "eval_duration": 1341630906
}

Notice that the model calls the correct tool but the tool requested is mentioned in content instead of tool_calls
This issue doesn't come when the user query is Fetch all books instead of List all books

Relevant log output

OS

Linux

GPU

No response

CPU

No response

Ollama version

0.11.8

Originally created by @docjon09 on GitHub (Sep 6, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/12200 ### What is the issue? When I run the api request with the following body ``` { "model": "hermes-3-8b:latest", "messages": [ { "role": "user", "content": "List all books" } ], "tools": [ { "type": "function", "function": { "name": "fetch_product_by_entity_code_and_entity_key_code", "description": "Fetch the product with given entity code and given entity key code", "parameters": { "type": "object", "properties": { "entityCode": { "type": "string", "description": "The entity code of the product" }, "entityKeyCode": { "type": "string", "description": "The entity key code of the product" } }, "required": ["entityCode", "entityKeyCode"] } } }, { "type": "function", "function": { "name": "fetch_all_books", "description": "Fetch all the books", "parameters": {} } } ], "options": { "temperature": 0 }, "stream": false, } ``` The output is as follows ``` { "model": "hermes-3-8b:latest", "created_at": "2025-09-06T10:53:48.80640778Z", "message": { "role": "assistant", "content": "[\n {\"name\":\"fetch_all_books\",\"arguments\":{}}\n]" }, "done_reason": "stop", "done": true, "total_duration": 8895005988, "load_duration": 70760242, "prompt_eval_count": 266, "prompt_eval_duration": 7481875206, "eval_count": 15, "eval_duration": 1341630906 } ``` Notice that the model calls the correct tool but the tool requested is mentioned in content instead of tool_calls This issue doesn't come when the user query is **Fetch all books** instead of **List all books** ### Relevant log output ```shell ``` ### OS Linux ### GPU _No response_ ### CPU _No response_ ### Ollama version 0.11.8
GiteaMirror added the bug label 2026-04-22 17:00:57 -05:00
GiteaMirror commented 2026-04-22 17:00:59 -05:00
Author
Owner
Copy Link

@docjon09 commented on GitHub (Sep 6, 2025):

The model used is hermes-3-8b

<!-- gh-comment-id:3261863367 --> @docjon09 commented on GitHub (Sep 6, 2025): The model used is hermes-3-8b
GiteaMirror commented 2026-04-22 17:00:59 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Sep 6, 2025):

Where did you get the model from?

<!-- gh-comment-id:3261882160 --> @rick-github commented on GitHub (Sep 6, 2025): Where did you get the model from?
GiteaMirror commented 2026-04-22 17:01:01 -05:00
Author
Owner
Copy Link

@docjon09 commented on GitHub (Sep 6, 2025):

From the ollama library https://ollama.com/library/hermes3:8b

<!-- gh-comment-id:3261903724 --> @docjon09 commented on GitHub (Sep 6, 2025): From the ollama library https://ollama.com/library/hermes3:8b
GiteaMirror commented 2026-04-22 17:01:02 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Sep 6, 2025):

https://ollama.com/library/hermes3:8b is not the same as hermes-3-8b. At the minimum the model has been renamed and possibly modified. If I use the model from the ollama library, your JSON request works fine.

for i in {1..10} ; do curl -s localhost:11434/api/chat -d @12200.json  | jq -c .message.tool_calls ; done | sort | uniq -c
     10 [{"function":{"name":"fetch_all_books","arguments":{}}}]
<!-- gh-comment-id:3261909473 --> @rick-github commented on GitHub (Sep 6, 2025): https://ollama.com/library/hermes3:8b is not the same as hermes-3-8b. At the minimum the model has been renamed and possibly modified. If I use the model from the ollama library, your JSON request works fine. ```console for i in {1..10} ; do curl -s localhost:11434/api/chat -d @12200.json | jq -c .message.tool_calls ; done | sort | uniq -c 10 [{"function":{"name":"fetch_all_books","arguments":{}}}] ```
GiteaMirror commented 2026-04-22 17:01:03 -05:00
Author
Owner
Copy Link

@docjon09 commented on GitHub (Sep 6, 2025):

Hi @rick-github , you were right. This wasn't downloaded from the library but from ModelFile and Meta-Llama-3.1-8B-Instruct-Q5_K_M.gguf file (a colleague did the installation so I wasn't aware). I will try with the hermes3:8b model from the library. Thanks a lot for the quick reply!

<!-- gh-comment-id:3262335757 --> @docjon09 commented on GitHub (Sep 6, 2025): Hi @rick-github , you were right. This wasn't downloaded from the library but from ModelFile and Meta-Llama-3.1-8B-Instruct-Q5_K_M.gguf file (a colleague did the installation so I wasn't aware). I will try with the hermes3:8b model from the library. Thanks a lot for the quick reply!
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#33875
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?