File tree Expand file tree Collapse file tree 2 files changed +4
-3
lines changed
Expand file tree Collapse file tree 2 files changed +4
-3
lines changed Original file line number Diff line number Diff line change @@ -126,7 +126,8 @@ def _build_server_command(self) -> List[str]:
126126
127127 # Add optimization flags
128128 if self .config ['enable_speculative_decode' ]:
129- cmd .extend (['--speculative-algorithm' , self .config ['speculative_algorithm' ]])
129+ cmd .extend (['--speculative-algorithm' ,
130+ self .config ['speculative_algorithm' ]])
130131 cmd .extend (['--speculative-num-steps' ,
131132 str (self .config ['speculative_num_steps' ])])
132133 cmd .extend (['--speculative-eagle-topk' ,
Original file line number Diff line number Diff line change 117117
118118 # True: enable with DS-R1 interactive scenario MTP config
119119 'enable_speculative_decode' : True ,
120- 'speculative_algorithm' : 'EAGLE' , # EAGLE1/2 style decoding with DS-R1 MTP Head
120+ 'speculative_algorithm' : 'EAGLE' , # EAGLE1/2 style decoding with DS-R1 MTP Head
121121 'speculative_num_steps' : 3 ,
122- 'speculative_topk' : 1 , # Linear (no draft-trees)
122+ 'speculative_topk' : 1 , # Linear (no draft-trees)
123123 },
124124 'env_vars' : {
125125 'CUDA_MODULE_LOADING' : 'LAZY' ,
You can’t perform that action at this time.
0 commit comments