refactor: :construction: Initial, not tested, refactor of LLM (#448)

* refactor: :construction: Initial, not tested, refactor of LLM

* refactor: :construction: replace usages of _complete with complete

* fix: :bug: fixes after refactor

* refactor: :recycle: template raw completions in chat format

* test: :white_check_mark: simplified edit prompt and UNIT TESTS!

* ci: :green_heart: unit tests in ci

* fix: :bug: fixes for unit tests in ci

* fix: :bug: start uvicorn in tests without poetry

* fix: :closed_lock_with_key: add secrets to main.yaml

* feat: :adhesive_bandage: timeout for all LLM classes

* ci: :green_heart: prepare main.yaml for main branch

3 files changed, 3 insertions, 3 deletions

diff --git a/docs/docs/concepts/sdk.md b/docs/docs/concepts/sdk.md
index ac2bc8bf..21190aa8 100644
--- a/docs/docs/concepts/sdk.md
+++ b/docs/docs/concepts/sdk.md
@@ -23,7 +23,7 @@ The **Continue SDK** gives you all the tools you need to automate software devel
 

 ### `sdk.models`

 

-`sdk.models` is an instance of the `Models` class, containing many of the most commonly used LLMs or other foundation models. You can access a model (starcoder for example) like `starcoder = sdk.models.starcoder`. Right now, all of the models are `LLM`s, meaning that they offer the `complete` method, used like `bubble_sort_code = await starcoder._complete("# Write a bubble sort function below, in Python:\n")`.

+`sdk.models` is an instance of the `Models` class, containing many of the most commonly used LLMs or other foundation models. You can access a model (starcoder for example) like `starcoder = sdk.models.starcoder`. Right now, all of the models are `LLM`s, meaning that they offer the `complete` method, used like `bubble_sort_code = await starcoder.complete("# Write a bubble sort function below, in Python:\n")`.

 

 ### `sdk.history`

 

diff --git a/docs/docs/customization.md b/docs/docs/customization.md
index 09f7ed46..2d1d8ba4 100644
--- a/docs/docs/customization.md
+++ b/docs/docs/customization.md
@@ -292,7 +292,7 @@ class CommitMessageStep(Step):
 
         # Ask the LLM to write a commit message,
         # and set it as the description of this step
-        self.description = await sdk.models.default._complete(
+        self.description = await sdk.models.default.complete(
             f"{diff}\n\nWrite a short, specific (less than 50 chars) commit message about the above changes:")
 
 config=ContinueConfig(
diff --git a/docs/docs/walkthroughs/create-a-recipe.md b/docs/docs/walkthroughs/create-a-recipe.md
index cc80be0e..2cb28f77 100644
--- a/docs/docs/walkthroughs/create-a-recipe.md
+++ b/docs/docs/walkthroughs/create-a-recipe.md
@@ -31,7 +31,7 @@ If you'd like to override the default description of your steps, which is just t
 

 - Return a static string

 - Store state in a class attribute (prepend with a double underscore, which signifies (through Pydantic) that this is not a parameter for the Step, just internal state) during the run method, and then grab this in the describe method.

-- Use state in conjunction with the `models` parameter of the describe method to autogenerate a description with a language model. For example, if you'd used an attribute called `__code_written` to store a string representing some code that was written, you could implement describe as `return models.medium._complete(f"{self.\_\_code_written}\n\nSummarize the changes made in the above code.")`.

+- Use state in conjunction with the `models` parameter of the describe method to autogenerate a description with a language model. For example, if you'd used an attribute called `__code_written` to store a string representing some code that was written, you could implement describe as `return models.medium.complete(f"{self.\_\_code_written}\n\nSummarize the changes made in the above code.")`.

 

 ## 2. Compose steps together into a complete recipe