Skip to content

Commit 8a96432

Browse files
authored
add glm benchmark yaml (#4289)
1 parent 67e693b commit 8a96432

File tree

3 files changed

+19
-0
lines changed

3 files changed

+19
-0
lines changed
Lines changed: 5 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,5 @@
1+
max_model_len: 32768
2+
max_num_seqs: 128
3+
tensor_parallel_size: 4
4+
use_cudagraph: True
5+
load_choices: "default_v1"
Lines changed: 6 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,6 @@
1+
max_model_len: 32768
2+
max_num_seqs: 128
3+
tensor_parallel_size: 4
4+
use_cudagraph: True
5+
load_choices: "default_v1"
6+
quantization: wfp8afp8
Lines changed: 8 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,8 @@
1+
top_p: 0.95
2+
temperature: 0.6
3+
metadata:
4+
min_tokens: 1
5+
max_tokens: 12288
6+
repetition_penalty: 1.0
7+
frequency_penalty: 0
8+
presence_penalty: 0

0 commit comments

Comments
 (0)