Repository navigation
docs(models): glm5.3 is served as NVFP4 and takes image, audio and video - #137
Merged
Merged
Conversation
The premium model moved to an NVFP4 build that is multimodal: text, image, audio and video in. Every surface still described the previous one, FP8 and text only, which #39 had set on purpose when the vision copied from the announcement was wrong. - The card (EN/ES) and the home table: NVFP4, the four input modalities and their bullets. - The model catalog gains a `video` modality, and glm5.3 is the first model to carry it. Choose a model gets a row for handing it a video. - The OpenAPI spec: the catalog row lists the inputs, and the messages field no longer singles glm5.3 out as text only. - The client configs: OpenCode declares all four (its schema has `video`), VS Code and Cline mark it with images. Pi's schema only has text and image, so it goes in with those and the page says so, as it already did for mimo-v2.6-flash. - The prose that steered readers away from it for images (Choose a model, Claude Code, Hermes, VS Code) goes. The two tests that asserted text only now assert the opposite, and were seen failing against the old pages. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
The premium
glm5.3moved to an NVFP4 build that is multimodal (text, image, audio and video in). The docs still said FP8 and text only, as #39 had set them.What changes (EN and ES)
/docs/modelscard and the home table: NVFP4,text · image · audio · video.modelCatalog.tsgains avideomodality, and glm5.3 is the first model with it. Choose a model gets a "hand the model a video" row.messagesfield no longer call it text only.video). VS Code and Cline mark it with images. Pi only accepts text and image, so it goes in with those and the known issues say so, like mimo-v2.6-flash.Tests
docsGlm53.test.tsandmodelsTable.test.tsnow require NVFP4 and the new modalities. They fail against the old pages and pass now.vitest: 1325/1325.npm run buildpasses.Not in this PR
nan-cliwrites the same tool configs and still declares glm5.3 as text only.nan/glm5.3) is maintained outside this repo.🤖 Generated with Claude Code