[Bugfix] Add missing attributes in mistral tokenizer

[DarkLight1337]

FIX (partial) #8358

Also add a warning to address https://github.com/vllm-project/vllm/pull/8314#discussion_r1751721812

cc @patrickvonplaten

[github-actions[bot]]

👋 Hi! Thank you for contributing to the vLLM project. Just a reminder: PRs would not trigger full CI run by default. Instead, it would only run fastcheck CI which starts running only a small and essential subset of CI tests to quickly catch errors. You can run other CI tests on top of those by going to your fastcheck build on Buildkite UI (linked in the PR checks section) and unblock them. If you do not have permission to unblock, ping simon-mo or khluu to add you in our Buildkite org.

Once the PR is approved and ready to go, your PR reviewer(s) can run CI to test the changes comprehensively before merging.

To run CI, PR reviewers can do one of these:

• Add ready label to the PR
• Enable auto-merge.

🚀

[ywang96]

I’m curious: is mistral tokenizer supposed to work with guidelines? I’m asking because I’m wondering if there’s a better way to patch this.

[DarkLight1337]

I’m curious: is mistral tokenizer supposed to work with guidelines? I’m asking because I’m wondering if there’s a better way to patch this.

Oh, I only saw the first part of the issue about the AttributeError - seems that the OP edited their message afterwards. Not sure about whether it can work with the various guided decoding backends.

[DarkLight1337]

Not sure about whether it can work with the various guided decoding backends.

We can address this in another PR. Let's get this merged first to limit the scope of this PR and avoid introducing conflicts.