fix(llm): parse config.json with JSON5 when deriving max context length - #11511
Merged
nv-nmailhot merged 2 commits intoJul 11, 2026
Merged
Conversation
…th (#11445) Signed-off-by: ssojrani@nvidia.com <ssojrani@nvidia.com> Signed-off-by: Ryan McCormick <rmccormick@nvidia.com>
![devin-ai-integration[bot]](https://avatars.githubusercontent.com/in/811515?s=60&v=4){kind=small}
![dynamo-review-agent[bot]](https://avatars.githubusercontent.com/u/201626793?s=60&v=4){kind=small}
Contributor
Author
|
vLLM CI failures appear to be a pre-existing failure introduced on another PR I think this PR might've broken something, but for some reason the filters on github CI triggers are wrong and this change didn't trigger vllm tests: #11374 |
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
3 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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
release/1.3.0.config.jsonwith JSON5 when deriving maximum context length, so unrelated non-finite literals such asInfinityno longer prevent model registration.text_config→ tokenizer fallback order and per-field error behavior.Fixes DIS-2378
Root cause
The max-context probe pre-parsed the entire model config with strict
serde_json. A bare non-finite literal anywhere in the file caused parsing to fail before Dynamo could read the relevantmax_position_embeddingsfield.Original PR
80c4cf56c9ee487d0da641c9fafab4053ff4d73a9c13be8459dc7e631216bad4ff4283cc96abc8faTest plan
cargo test --locked -p dynamo-llm --lib --no-default-features model_card::ownership_tests::architectural_context_(3 passed)