error: [handleAbortError] AI response error; aborting request: Config not found for the ollama custom endpoint.

[lexciobotariu]

What happened?

I'm getting this error when I run the app with ollama separately.

This is how my librechat.yaml looks like for the ollama model

endpoints:
  custom:
    - name: "Ollama"
      apiKey: "ollama"
      # use 'host.docker.internal' instead of localhost if running LibreChat in a docker container
      baseURL: "http://127.0.0.1:11434/v1/chat/completions" 
      models:
        default: [
          "llama2",
          "mistral",
          "codellama",
          "dolphin-mixtral",
          "mistral-openorca"
          ]
        # fetching list of models is supported but the `name` field must start
        # with `ollama` (case-insensitive), as it does in this example.
        fetch: true
      titleConvo: true
      titleModel: "current_model"
      summarize: false
      summaryModel: "current_model"
      forcePrompt: true
      modelDisplayLabel: "Ollama"
      addParams:
          "stop": [
              "<|start_header_id|>",
              "<|end_header_id|>",
              "<|eot_id|>",
              "<|reserved_special_token"
          ]

When I access the chat it finds the model I have installed, but when I type a message I get this error

Something went wrong. Here's the specific error message we encountered: An error occurred while processing your request. Please contact the Admin.

Also I see in the vscode console warn: Invalid conversation ID: null I added the whole error down below.

I checked the indentation of the YAML file, and it seems correct.

Steps to Reproduce

1. Install Ollama, download any model
2. Configure librachat.yaml as above.
3. Run the project with NPM

What browsers are you seeing the problem on?

No response

Relevant log output

2024-11-08 16:55:53 error: [handleAbortError] AI response error; aborting request: Config not found for the ollama custom endpoint.
2024-11-08 16:55:53 warn: Invalid conversation ID: null
2024-11-08 16:55:53 info: ---`saveMessage` context: undefined
2024-11-08 16:55:53 info: ---Invalid conversation ID Params: {
  "sender": "Ollama",
  "messageId": "aa6c5bcf-a813-46a6-a37b-db60c7f000eb",
  "conversationId": null,
  "parentMessageId": "00000000-0000-0000-0000-000000000000",
  "unfinished": false,
  "error": true,
  "final": true,
  "text": "An error occurred while processing your request. Please contact the Admin.",
  "isCreatedByUser": false,
  "user": "672df0d453e07c7ddd08eafa"
}

Screenshots

No response

Code of Conduct

• I agree to follow this project's Code of Conduct

[m4druga]

I'm dealing with the same issue.

[danny-avila]

Try updating to the latest build, this issue was resolved

[m4druga]

I didn't find the root cause, but I found a way to work around the problem: the value of name needs to be in lowercase.

endpoints:
custom:
- name: "ollama"
apiKey: "ollama"

This is new, as previous deployments used name with only the first letter in uppercase without causing this error. Interestingly, the example on the LibreChat documentation site still shows name with an uppercase 'O'.

[daSonne]

this is driving crazy
running liberchat + ollama on docker (mac)

(librechat.yaml)
custom:
- name: "ollama"
apiKey: "ollama"
# use 'host.docker.internal' instead of localhost if running LibreChat in a docker container
baseURL: "http://host.docker.internal:11434/v1/"
models:
default: [
"llama2",
"mistral",
"codellama",
"dolphin-mixtral",
"mistral-openorca"
]
# fetching list of models is supported but the name field must start
# with ollama (case-insensitive), as it does in this example.
fetch: true
titleConvo: true
titleModel: "current_model"
summarize: false
summaryModel: "current_model"
forcePrompt: false
modelDisplayLabel: "Ollama"
I tried every solution I found no luck !
help!

[himefei]

Hi guys, for those who still can't get it to work, and you have already tried updating, change Ollama to lowercase. Please try #2745, it worked out for me by disableing Meli search

[danny-avila]

Apologies, it seems I missed something in my update, it should be fixed after this is merged and to remain case-insensitive:

#4681