readme : model : mtdm : lfm2 improvements #15476
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
CISC
approved these changes
Aug 21, 2025
| output = create_tensor(tn(LLM_TENSOR_OUTPUT, "weight"), {n_embd, n_vocab}, TENSOR_NOT_REQUIRED); | ||
|
|
||
| if (output == NULL) { | ||
| output = create_tensor(tn(LLM_TENSOR_TOKEN_EMBD, "weight"), {n_embd, n_vocab}, TENSOR_DUPLICATED); |
There was a problem hiding this comment.
You forgot to update llm_build_lfm2, it's still using model.tok_embd as output.
There was a problem hiding this comment.
very nice catch, probably got lost during preparing the upstream version, fixed in 9a8714b, thank you!
qnixsynapse
pushed a commit
to menloresearch/llama.cpp
that referenced
this pull request
Aug 25, 2025
* Support untied embeddings * Increase number of image tokens to 1024 * Add LFM2-VL to readme * Actually use untied embeddings
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Outdated suggestions cannot be applied.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.
Batched minor improvements into the
LFM2family into a single PRLFM2-VLinto readmeLFM2-VLto 1024 to improve OCRLFM2dense to unlock conversion of PTQ quantized checkpoints with untied embeddings