Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Cannot change role verifier and auto-fix to custom model. No model settings were updated #171

Closed
schniggie opened this issue Jul 14, 2024 · 3 comments
Assignees

Comments

@schniggie
Copy link

Trying to set my local ollama/deepseek-coder-v2 for all roles.

plandex models default
+---------------------------------+   
| Org-Wide Default Model Settings |
+---------------------------------+

�️  Current Model Set
+------------------------------------------------------------------+
|                          gpt-4o-latest                           |
+------------------------------------------------------------------+
| Uses OpenAI's latest gpt-4o model, first released on 2024-05-13, |
| for heavy lifting, and latest version of gpt-3.5-turbo for       |
| lighter tasks.                                                   |
+------------------------------------------------------------------+

� Models
+-----------------+----------+-------------------+-------------+-------+
|      ROLE       | PROVIDER |       MODEL       | TEMPERATURE | TOP P |
+-----------------+----------+-------------------+-------------+-------+
| planner         | custom   | deepseek-coder-v2 |         0.3 |   0.3 |
| summarizer      | custom   | deepseek-coder-v2 |         0.2 |   0.2 |
| builder         | custom   | deepseek-coder-v2 |         0.1 |   0.1 |
| names           | custom   | deepseek-coder-v2 |         0.8 |   0.5 |
| commit-messages | custom   | deepseek-coder-v2 |         0.8 |   0.5 |
| auto-continue   | custom   | deepseek-coder-v2 |         0.1 |   0.1 |
| verifier        | openai   | gpt-4o            |         0.2 |   0.2 |
| auto-fix        | openai   | gpt-4o            |         0.1 |   0.1 |
+-----------------+----------+-------------------+-------------+-------+

� Planner Defaults
+------------+------------------+------------------------+
| MAX TOKENS | MAX CONVO TOKENS | RESERVED OUTPUT TOKENS |
+------------+------------------+------------------------+
|     128000 |            10000 |                   4000 |
+------------+------------------+------------------------+

When I try again to change the default model for verifier and auto-fix it fails:

plandex set-model default
? Choose a new model pack, or select a role or override to update: 🤖 role | verifier → verifies file correctness
? Select a property to update: Select a model
? Select a provider: ollama
? Select a model: ollama → deepseek-coder-v2 | max 128000 🪙
🤷‍♂️ No model settings were updated

Same goes for auto-fix. This also affects setting the models for the current plan. Any idea?

@ryzen3100
Copy link

I have experienced the same bug.
If you create a new model-pack by running plandex model-packs create and then setting the new model-pack it should work.

PS.: Have you had any success setting up deepseek-coder? I could not get it work at all using openrouter.

@danenania
Copy link
Contributor

Thanks for reporting! This will be fixed soon.

@danenania danenania assigned danenania and unassigned danenania Jul 19, 2024
@danenania danenania added the good first issue Good for newcomers label Jul 23, 2024
@danenania danenania removed their assignment Jul 23, 2024
@danenania danenania self-assigned this Aug 15, 2024
@danenania
Copy link
Contributor

This is done and will be in the next release.

@danenania danenania removed the good first issue Good for newcomers label Aug 15, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

3 participants