Add DeepseekV4 Flash support: inference, converter, MTP speculative decoding#959
Open
iStrzalka wants to merge 1 commit into
Open
Add DeepseekV4 Flash support: inference, converter, MTP speculative decoding#959iStrzalka wants to merge 1 commit into
iStrzalka wants to merge 1 commit into
Conversation
|
Thanks for your pull request! It looks like this may be your first contribution to a Google open source project. Before we can look at your pull request, you'll need to sign a Contributor License Agreement (CLA). View this failed invocation of the CLA check for more information. For the most up to date status, view the checks section at the bottom of the pull request. |
jan-wassenberg
requested changes
Jul 20, 2026
jan-wassenberg
left a comment
jan-wassenberg
left a comment
Member
There was a problem hiding this comment.
Interesting, thanks for adding! Excellent work.
Looks like we have some merge conflicts in kv_cache.cc, gemma.cc, configs.h/cc, weights.h, python/configs.cc - please resolve them to unblock merge. They seem independent, we can keep both.
| return config; | ||
| } | ||
|
|
||
| <<<<<<< ours |
| GEMMA3_12B_LM, | ||
| GEMMA3_27B_LM, | ||
| GEMMA4_26B_MOE, | ||
| <<<<<<< ours |
|
|
||
| InternalModelConfig internal; | ||
| bool use_global_timescale = false; // for Gemma 3 | ||
| <<<<<<< ours |
|
|
||
| // Fill activations.x (much faster than TransformerLayer). | ||
| size_t image_token_position = 0; | ||
| <<<<<<< ours |
| // fires. The 2-arg constructor path relies on k_v_cols == 0 to skip reshape. | ||
| } | ||
|
|
||
| <<<<<<< ours |
iStrzalka
force-pushed
the
deepseek-v4-flash
branch
from
July 20, 2026 15:48
fb7d1bb to
27cf8cb
Compare
…ecoding. Tested via running the model to check whether output made sense.
iStrzalka
force-pushed
the
deepseek-v4-flash
branch
from
July 20, 2026 16:19
27cf8cb to
2895479
Compare
jan-wassenberg
added
the
copybara-import
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
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.
Adds DeepseekV4 Flash support: inference, converter, and MTP speculative decoding.
Tested via running the model to check whether output made sense.