Skip to content
GitLab
Menu
Projects
Groups
Snippets
Loading...
Help
Help
Support
Community forum
Keyboard shortcuts
?
Submit feedback
Contribute to GitLab
Sign in
Toggle navigation
Menu
Open sidebar
OpenDAS
vllm_cscc
Commits
dad2d6a5
Unverified
Commit
dad2d6a5
authored
Feb 03, 2026
by
Lucas Hänke de Cansino
Committed by
GitHub
Feb 03, 2026
Browse files
[Bugfix][Model] Fix DeepSeek-OCR-2 chat template to include BOS token (#33642)
Signed-off-by:
l4b4r4b4b4
<
lucas.cansino@mail.de
>
parent
32e84fa1
Changes
1
Hide whitespace changes
Inline
Side-by-side
Showing
1 changed file
with
5 additions
and
4 deletions
+5
-4
vllm/transformers_utils/configs/deepseek_vl2.py
vllm/transformers_utils/configs/deepseek_vl2.py
+5
-4
No files found.
vllm/transformers_utils/configs/deepseek_vl2.py
View file @
dad2d6a5
...
...
@@ -119,8 +119,9 @@ class DeepseekVLV2Config(PretrainedConfig):
self
.
candidate_resolutions
=
candidate_resolutions
self
.
vocab_size
=
self
.
text_config
.
vocab_size
# update model_type for OCR model
if
"DeepseekOCRForCausalLM"
in
(
self
.
architectures
or
kwargs
.
get
(
"architectures"
,
[])
):
# update model_type for OCR models
architectures
=
self
.
architectures
or
kwargs
.
get
(
"architectures"
,
[])
if
"DeepseekOCRForCausalLM"
in
architectures
:
self
.
model_type
=
"deepseek_ocr"
elif
"DeepseekOCR2ForCausalLM"
in
architectures
:
self
.
model_type
=
"deepseek_ocr2"
Write
Preview
Markdown
is supported
0%
Try again
or
attach a new file
.
Attach a file
Cancel
You are about to add
0
people
to the discussion. Proceed with caution.
Finish editing this message first!
Cancel
Please
register
or
sign in
to comment