Gemini Flash response only 2048 character and not continues code generation.

[robtail]

Before submitting your bug report

• [X] I believe this is a bug. I'll try to join the Continue Discord for questions
• [X] I'm not able to find an open issue that reports the same bug
• [X] I've seen the troubleshooting guide on the Continue Docs

Relevant environment info

- OS: win 11
- Continue: latest - v0.8.43
- IDE: VSCode
- Model: Gemini 1.5 Flash
- config.json:
  {
  "models": [
    {
      "title": "llama3:8b-instruct",
      "provider": "ollama",
      "model": "llama3:8b-instruct-q6_K",
      "temperature": "0.3"
    },
    {
      "title": "Gemini 1.5 Flash",
      "model": "gemini-1.5-flash-latest",
      "contextLength": 1000000,
      "max_output_tokens": 8192,
      "temperature": 0.5,
      "apiKey": "Mykey",
      "provider": "gemini"
    }
  ],
  "customCommands": [
    {
      "name": "test",
      "prompt": "{{{ input }}}\n\nWrite a comprehensive set of unit tests for the selected code. It should setup, run tests that check for correctness including important edge cases, and teardown. Ensure that the tests are complete and sophisticated. Give the tests just as chat output, don't edit any file.",
      "description": "Write unit tests for highlighted code"
    }
  ],
  "tabAutocompleteModel": {
    "title": "deepseek-coder:6.7b-base",
    "provider": "ollama",
    "model": "deepseek-coder:6.7b-base-q6_K"
  },
  "embeddingsProvider": {
    "provider": "ollama",
    "model": "nomic-embed-text"
  }
}

### Description

Hi,

Using Gemini Flash 1.5 model. 
In chat asking to generate a "long" program code (snake game in python.) 
It starts generate the code but stops generate at somewhere. 
I'm pushing  "Continue generation" button but instead to continue the code it starts generating the code again and stops at the same line again. 

How to increase response length? or how force to continue?

This is the prompt issued to continue code generation:
Continue your response exactly where you left off:

### To reproduce

_No response_

### Log output

_No response_

[sestinj]

@robtail You can use the "contextLength" option in your model in config.json to set a longer contextLength, or "completionOptions.maxTokens" for the output length. We'll be doing a better job of automatically detecting this in upcoming releases

[robtail]

Hi Sestinj,

Thanks for the fast reply! :) contextLength was already set to million, see the original bug report: "contextLength": 1000000, And now i tried completionOptions.maxTokens to
"completionOptions.maxTokens": 8192,

but same issue still, it stop generate output after 2054 character. :( and the continue button not continue but start code generate again.

any suggestion?