Added instructions for configuring Ollama models

[thecatontheflat]

Important

Added detailed instructions for configuring Ollama models, including context size adjustment and updated Roo Code setup.

• Instructions for Ollama:
  • Added command ollama serve to ensure Ollama is running.
  • Changed model download command from ollama run to ollama pull.
  • Added new model qwen2.5-coder:32b to recommended list.
  • Added steps to configure model context size to 32k tokens using /set parameter num_ctx 32768 and save with /save your_model_name.
• Roo Code Configuration:
  • Updated instructions to use the new model name after saving.
  • Added optional configuration for model context size in advanced settings.

^{This description was created by for bf21e2b. It will automatically update as commits are pushed.}

[ellipsis-dev]

❌ Changes requested. Reviewed everything up to bf21e2b in 1 minute and 10 seconds

More details

• Looked at 83 lines of code in 1 files
• Skipped 0 files when reviewing.
• Skipped posting 8 drafted comments based on config settings.

1. docs/advanced-usage/local-models.md:26

• Draft comment:
  Consider adding a period at the end of 'Make sure Ollama is running' for consistency with other instructions.
• Reason this comment was not posted:
  Comment did not seem useful. Confidence is useful = 0% <= threshold 50%
  This comment is purely about punctuation and does not address any functional or structural aspect of the code. It does not align with the rules for useful comments, which should focus on code functionality, structure, or adherence to development standards.

2. docs/advanced-usage/local-models.md:45

• Draft comment:
  Verify that 'ollama pull' is the correct command for downloading models and update related instructions if necessary.
• Reason this comment was not posted:
  Marked as duplicate.

3. docs/advanced-usage/local-models.md:70

• Draft comment:
  Consider adding a note on how to exit the model session (e.g., using '/bye') after saving the model, to provide full guidance.
• Reason this comment was not posted:
  Comment did not seem useful. Confidence is useful = 0% <= threshold 50%
  This comment is purely informative and suggests adding a note for user guidance, which is not necessary for code review. It doesn't align with the rules of not making purely informative comments.

4. docs/advanced-usage/local-models.md:26

• Draft comment:
  Good addition: reminding users to run 'ollama serve' before other operations. This ensures Ollama is running as needed.
• Reason this comment was not posted:
  Confidence changes required: 0% <= threshold 50%
  None

5. docs/advanced-usage/local-models.md:45

• Draft comment:
  Changed command from 'ollama run' to 'ollama pull' for model download. Confirm that 'pull' is intentional for downloading without launching.
• Reason this comment was not posted:
  Confidence changes required: 0% <= threshold 50%
  None

6. docs/advanced-usage/local-models.md:37

• Draft comment:
  Added 'qwen2.5-coder:32b' model recommendation. Consider a brief note on its advantages for coding tasks.
• Reason this comment was not posted:
  Confidence changes required: 0% <= threshold 50%
  None

7. docs/advanced-usage/local-models.md:80

• Draft comment:
  Typo: 'Advance settings' should be updated to 'Advanced settings' for consistency and clarity.
• Reason this comment was not posted:
  Confidence changes required: 0% <= threshold 50%
  None

8. docs/advanced-usage/local-models.md:64

• Draft comment:
  Consider adding a note after saving the model, e.g., reminding users to exit the model session with '/bye'.
• Reason this comment was not posted:
  Confidence changes required: 33% <= threshold 50%
  None

Workflow ID: wflow_B3vjmrCOL8NBTU8y

---

Want Ellipsis to fix these issues? Tag @ellipsis-dev in a comment. You can customize Ellipsis with 👍 / 👎 feedback, review rules, user-specific overrides, quiet mode, and more.

[mrubens]

Thank you!