Fix: GBNF missing "root" node crashing server #6004
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
The fix should be in |
…urning valid grammar structure. This is an alternate fix location to put the required root node in sampling.cpp instead of grammar-parser.cpp.
HanClinto
force-pushed
the
fix_gbnf_root_required
branch
from
bfcf545 to
10ac60b
Compare
Excellent, thank you! This branch is now updated. |
ggerganov
approved these changes
Mar 13, 2024
NeoZhangJianyu
pushed a commit
to NeoZhangJianyu/llama.cpp
that referenced
this pull request
Mar 15, 2024
Merged
hodlen
pushed a commit
to hodlen/llama.cpp
that referenced
this pull request
Apr 1, 2024
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Resolves #3878 by ensuring the existence of a "root" symbol before returning a valid grammar.
NOTE: There are two possible locations to put this fix -- the alternate location is to put it in
sampling.cpp->llama_sampling_init(), which is the only place where the "root" node is named as required. That fix would look like this.@ggerganov Do you have preference on the two versions of this fix? I think this main difference is -- philosophically -- do we want all grammars to require a "root" node (if so, then put it in grammar-parser), or if it's possible to have a grammar without a "root" node (that's technically a syntactically correct grammar, but it's missing our expected entry point), then we should put the fix in sampling.cpp.