Fix Harmony reasoning parser for and auto-separation for gpt-oss models (#9190)
Co-authored-by: Chang Su <chang.s.su@oracle.com> Co-authored-by: Chayenne <zhaochen20@outlook.com> Co-authored-by: zhaochenyang20 <zhaochenyang20@gmail.com> Co-authored-by: minleminzui <2969413251@qq.com> Co-authored-by: maocheng23 <maocheng@berkeley.edu> Co-authored-by: Xinyuan Tong <xinyuantong.cs@gmail.com>
This commit is contained in:
@@ -2271,6 +2271,7 @@ class ServerArgs:
|
||||
if is_mxfp4_quant_format:
|
||||
# use bf16 for mxfp4 triton kernels
|
||||
self.dtype = "bfloat16"
|
||||
|
||||
elif "Llama4" in model_arch:
|
||||
assert self.attention_backend in {
|
||||
"fa3",
|
||||
|
||||
Reference in New Issue
Block a user