Add Mistral3 multimodal support with Pixtral vision encoder #431
+1,785
−8
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
This adds support for Mistral3 multimodal models (vision + text): - `Bumblebee.Vision.Pixtral`: Pixtral vision encoder with RoPE support - `Bumblebee.Text.Mistral3`: Mistral3 text decoder with interleaved attention - `Bumblebee.Multimodal.Mistral3`: Vision-language model combining Pixtral and Mistral3 with multimodal projector for image-conditioned generation - Ministral/Ministral3 variant support with interleaved attention - Devstral 2 (Ministral3) model support Supported architectures: - PixtralVisionModel - Mistral3Model, Mistral3ForCausalLM, Mistral3ForSequenceClassification - Mistral3ForConditionalGeneration (multimodal) - Ministral3ForCausalLM
dai-yamashita
force-pushed
the
add-mistral3-multimodal-support
branch
from
dd11e25 to
911cc2d
Compare
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Outdated suggestions cannot be applied.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.
Summary
This PR adds support for Mistral3 multimodal models (Ministral series):
attention_head_size,use_interleaved_attention,tie_word_embeddingsBumblebee.Vision.PixtralBumblebee.Text.Mistral3Bumblebee.Multimodal.Mistral3Supported architectures
PixtralVisionModelMistral3Model,Mistral3ForCausalLM,Mistral3ForSequenceClassificationMistral3ForConditionalGeneration(multimodal)Key features
transformer.exfor per-layer attention configurationTest plan