Fix Hermes chat template validation error

[Br1an67]

Fixes #4150

Summary

Hermes models use ChatML-style templates that don't require the {% if add_generation_prompt %} validation check. This PR adds hermes to the list of models that bypass this check, similar to how mistral and qwen3guard are already handled.

Changes

• Added hermes to the model bypass list in load_correct_tokenizer function in unsloth/tokenizer_utils.py

Root Cause

When loading a Hermes model (or a LoRA adapter trained with Hermes/ChatML template), the fix_chat_template function was raising a RuntimeError because:

1. The ChatML template doesn't have {% if add_generation_prompt %} in a format the auto-fixer expects
2. The auto-fixer _fix_chat_template couldn't patch it automatically
3. This caused the error: "does not have a {% if add_generation_prompt %} for generation purposes"

Testing

The fix follows the existing pattern used for mistral and qwen3guard models, which have similar template structures that don't require this validation.

Diff Stats

 unsloth/tokenizer_utils.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

[gemini-code-assist]

Summary of Changes

Hello, I'm Gemini Code Assist¹! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed!

This pull request resolves a RuntimeError that occurred when loading Hermes models due to an incompatibility with a specific chat template validation check. By extending an existing bypass mechanism to include Hermes models, it ensures that models utilizing ChatML-style templates can be loaded without issues, thereby enhancing the robustness and compatibility of the tokenizer loading process.

Highlights

• Hermes Model Compatibility: Added "hermes" to the list of models that bypass the {% if add_generation_prompt %} validation check in load_correct_tokenizer to prevent RuntimeError when loading Hermes models with ChatML-style templates.

Changelog

• unsloth/tokenizer_utils.py
  • Added 'hermes' to the list of models that bypass the add_generation_prompt validation.

Activity

• No specific activity has been recorded for this pull request yet.

Using Gemini Code Assist

The full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips.

Invoking Gemini

You can request assistance from Gemini at any point by creating a comment using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands on the current page.

Feature	Command	Description
Code Review	/gemini review	Performs a code review for the current pull request in its current state.
Pull Request Summary	/gemini summary	Provides a summary of the current pull request in its current state.
Comment	@gemini-code-assist	Responds in comments when explicitly tagged, both in pull request comments and review comments.
Help	/gemini help	Displays a list of available commands.

Customization

To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a .gemini/ folder in the base of the repository. Detailed instructions can be found here.

Limitations & Feedback

Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for Github and other Google products, sign up here.

You can also get AI-powered code generation, chat, as well as code reviews directly in the IDE at no cost with the Gemini Code Assist IDE Extension.

Footnotes

1. Review the Privacy Notices, Generative AI Prohibited Use Policy, Terms of Service, and learn how to configure Gemini Code Assist in GitHub here. Gemini can make mistakes, so double check it and use code with caution. ↩

[gemini-code-assist]

Code Review

This pull request correctly adds hermes to the list of models that bypass the chat template validation, fixing a RuntimeError for Hermes models that use ChatML-style templates. The change is straightforward and follows the existing pattern for models like mistral. I've added one suggestion to improve the code comment for accuracy and to use a tuple for the list of models, which is a minor style improvement.

[gemini-code-assist]

The comment on line 604 is now a bit inaccurate. While mistral models might lack add_generation_prompt, other models in this list (like hermes which uses ChatML) do have it. The more general reason for bypassing is that their templates don't need the automatic fix from fix_chat_template.

For better code style and a minor performance improvement, it's good practice to use a tuple for a fixed collection of items instead of a list.

Suggested change

	# Ignore mistral type models since they don't have an add_generation_prompt
	if any(
	s in str(getattr(tokenizer, "name_or_path", "")).lower()
	for s in ["mistral", "qwen3guard"]
	for s in ["mistral", "qwen3guard", "hermes"]
	# Ignore some models that do not need chat template fixing.
	if any(
	s in str(getattr(tokenizer, "name_or_path", "")).lower()
	for s in ("mistral", "qwen3guard", "hermes")