Skip to content

model: correctly support input vision for deepseek4 - #28154

Open
ngxson wants to merge 2 commits into
xsn/dsv4_visionfrom
xsn/dsv4_vision_prob_bias
Open

model: correctly support input vision for deepseek4#28154
ngxson wants to merge 2 commits into
xsn/dsv4_visionfrom
xsn/dsv4_vision_prob_bias

Conversation

@ngxson

@ngxson ngxson commented Sep 1, 2026

Copy link
Copy Markdown
Collaborator

Overview

Stack on top of #28133

The DeepSeek-V4-Flash-Vision-Exp requires two extra things:

  • specific exp_probs bias for vision input
  • do not apply SWA when non_causal is set

Requirements

@ngxson ngxson mentioned this pull request Sep 1, 2026
Open
@github-actions github-actions Bot added model Model specific mtmd Related to multimodal functionality (video/image/audio) conversion labels Sep 1, 2026
@ngxson
ngxson marked this pull request as ready for review September 1, 2026 09:59
@ngxson
ngxson requested review from a team, CISC and ggerganov as code owners September 1, 2026 09:59
@tarruda tarruda mentioned this pull request Sep 1, 2026
Closed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@CISC CISC Awaiting requested review from CISC CISC is a code owner
@ggerganov ggerganov Awaiting requested review from ggerganov ggerganov is a code owner

Assignees

No one assigned

Labels

conversion model Model specific mtmd Related to multimodal functionality (video/image/audio)

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

1 participant

@ngxson