Support GLM tool calling and report CLI token-limit truncation - #66
Merged
Conversation
Integrate GLM tool protocol work from PR #61, validate rendering against the vendored template, preserve schema argument types, and reject unsupported tool history. Report token-limit exhaustion without mislabeling natural stops. Co-authored-by: hendrikras <3663335+hendrikras@users.noreply.github.com>
Closed
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
GLM chat containers currently reject tool definitions, and CLI replies can stop at the default 128-token limit without explaining why. This change enables GLM's native tool protocol and reports token-limit exhaustion with guidance to increase
-n, while preserving natural stop behavior.The GLM implementation incorporates #61 by @hendrikras, with additional fixes for template declaration fidelity, typed arguments, function-name whitespace, invalid tool definitions, unsupported tool history, and standalone stop-marker validation. It includes the upstream template and a strict CI oracle, HTTP/parser regression tests, and updated documentation.
Validation before rebasing onto current main:
make all test; synthetic full suite (61 passed, 0 failed, 14 prerequisite skips); server suite (272 tests, 3 prerequisite skips); final focused template/parser/chat-format suite (71 passed). GitHub CI will validate the rebased branch across the configured platforms. No end-to-end inference validation with real GLM weights.Addresses #60 and #59. Incorporates #61; issues will be closed after successful CI and merge.