Add new o1-mini and o1-preview models

[KennyDizi]

PR Type

Enhancement

---

Description

• Added new o1-mini and o1-mini-2024-09-12 models with a token limit of 128K.
• Added new o1-preview and o1-preview-2024-09-12 models with a token limit of 128K.

---

Changes walkthrough

	Relevant files
Enhancement	__init__.py Add new o1-mini and o1-preview models                                        pr_agent/algo/init.py Added new model entries o1-mini and o1-mini-2024-09-12. Added new model entries o1-preview and o1-preview-2024-09-12. +4/-0

[qodo-merge-pro]

PR-Agent was enabled for this repository. To continue using it, please link your git user with your CodiumAI identity here.

PR Reviewer Guide 🔍

⏱️ Estimated effort to review: 1 🔵⚪⚪⚪⚪

🏅 Score: 95

🧪 No relevant tests

🔒 No security concerns identified

🔀 No multiple PR themes

⚡ No key issues to review

[qodo-merge-pro]

PR-Agent was enabled for this repository. To continue using it, please link your git user with your CodiumAI identity here.

PR Code Suggestions ✨

Category	Suggestion	Score
Maintainability	Define constants for repeated values to improve maintainability Consider adding a constant for the token limit (128000) and the comment text to avoid repetition and make future updates easier. pr_agent/algo/init.py [22-25] -'o1-mini': 128000, # 128K, but may be limited by config.max_model_tokens -'o1-mini-2024-09-12': 128000, # 128K, but may be limited by config.max_model_tokens -'o1-preview': 128000, # 128K, but may be limited by config.max_model_tokens -'o1-preview-2024-09-12': 128000, # 128K, but may be limited by config.max_model_tokens +O1_TOKEN_LIMIT = 128000 +O1_COMMENT = "128K, but may be limited by config.max_model_tokens" +'o1-mini': O1_TOKEN_LIMIT, # O1_COMMENT +'o1-mini-2024-09-12': O1_TOKEN_LIMIT, # O1_COMMENT +'o1-preview': O1_TOKEN_LIMIT, # O1_COMMENT +'o1-preview-2024-09-12': O1_TOKEN_LIMIT, # O1_COMMENT Apply this suggestion Suggestion importance[1-10]: 8 Why: This suggestion improves code maintainability by reducing repetition and making future updates easier, which is particularly valuable for configuration data like token limits.	8
Enhancement	Use a dictionary comprehension to create entries for similar models Consider using a dictionary comprehension to create the entries for the 'o1' family models. This can make the code more concise and easier to maintain, especially if more models are added in the future. pr_agent/algo/init.py [22-25] -'o1-mini': 128000, # 128K, but may be limited by config.max_model_tokens -'o1-mini-2024-09-12': 128000, # 128K, but may be limited by config.max_model_tokens -'o1-preview': 128000, # 128K, but may be limited by config.max_model_tokens -'o1-preview-2024-09-12': 128000, # 128K, but may be limited by config.max_model_tokens +**o1_models = {f'o1-{model}{"-" + date if date else ""}': 128000 for model in ['mini', 'preview'] for date in ['', '2024-09-12']} Apply this suggestion Suggestion importance[1-10]: 7 Why: The suggestion offers a more concise and maintainable way to define similar model entries, which is beneficial for code readability and future updates.	7

[KennyDizi]

👑 🌏 I have synchronized the PR description with the latest commit (a944962).

[mrT23]

I will try to benchmark them on this
https://pr-agent-docs.codium.ai/finetuning_benchmark/
when I have the time