Skip to content
Projects
Groups
Snippets
Help
Loading...
Help
Contribute to GitLab
Sign in
Toggle navigation
C
coderai
Project
Project
Details
Activity
Cycle Analytics
Repository
Repository
Files
Commits
Branches
Tags
Contributors
Graph
Compare
Charts
Issues
0
Issues
0
List
Board
Labels
Milestones
Merge Requests
0
Merge Requests
0
CI / CD
CI / CD
Pipelines
Jobs
Schedules
Charts
Wiki
Wiki
Snippets
Snippets
Members
Members
Collapse sidebar
Close sidebar
Activity
Graph
Charts
Create a new issue
Jobs
Commits
Issue Boards
Open sidebar
nexlab
coderai
Commits
8665016a
Commit
8665016a
authored
Mar 01, 2026
by
Stefy Lanza (nextime / spora )
Browse files
Options
Browse Files
Download
Email Patches
Plain Diff
Add 'a3b' to MoE model indicators to recognize Qwen3.5-A3B architecture
parent
4041187d
Changes
1
Hide whitespace changes
Inline
Side-by-side
Showing
1 changed file
with
1 addition
and
1 deletion
+1
-1
coderai
coderai
+1
-1
No files found.
coderai
View file @
8665016a
...
...
@@ -589,7 +589,7 @@ class NvidiaBackend(ModelBackend):
def
_is_moe_model
(
self
,
model_name
:
str
)
->
bool
:
"""Check if model is a MoE (Mixture of Experts) model which needs more VRAM headroom."""
moe_indicators
=
[
'moe'
,
'mixtral'
,
'qwen3_5_moe'
,
'qwen3.5_moe'
,
'expert'
]
moe_indicators
=
[
'moe'
,
'mixtral'
,
'qwen3_5_moe'
,
'qwen3.5_moe'
,
'expert'
,
'a3b'
]
model_name_lower
=
model_name
.
lower
()
return
any
(
indicator
in
model_name_lower
for
indicator
in
moe_indicators
)
...
...
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