ruihan/qwen38-flash-next-rtx6000d
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Compare MTP steps and prefill batches; retain qualified MTP 2 baseline

Browse Source
This commit is contained in:
ruihan
2026-09-18 23:22:45 +08:00
parent 918b78a3b8
commit c8bfe6b9ab
45 changed files with 3785 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files
audit/tuning/steps-123/mtp1-b2048/mtp-metrics.txt
+24
View File
@@ -0,0 +1,24 @@
# HELP vllm:spec_decode_num_drafts_total Number of spec decoding drafts.
# TYPE vllm:spec_decode_num_drafts_total counter
vllm:spec_decode_num_drafts_total{engine="0",model_name="qwen3.8-flash-next"} 13297.0
# HELP vllm:spec_decode_num_drafts_created Number of spec decoding drafts.
# TYPE vllm:spec_decode_num_drafts_created gauge
vllm:spec_decode_num_drafts_created{engine="0",model_name="qwen3.8-flash-next"} 1.7897433982360141e+09
# HELP vllm:spec_decode_num_draft_tokens_total Number of draft tokens.
# TYPE vllm:spec_decode_num_draft_tokens_total counter
vllm:spec_decode_num_draft_tokens_total{engine="0",model_name="qwen3.8-flash-next"} 13297.0
# HELP vllm:spec_decode_num_draft_tokens_created Number of draft tokens.
# TYPE vllm:spec_decode_num_draft_tokens_created gauge
vllm:spec_decode_num_draft_tokens_created{engine="0",model_name="qwen3.8-flash-next"} 1.7897433982360315e+09
# HELP vllm:spec_decode_num_accepted_tokens_total Number of accepted tokens.
# TYPE vllm:spec_decode_num_accepted_tokens_total counter
vllm:spec_decode_num_accepted_tokens_total{engine="0",model_name="qwen3.8-flash-next"} 10374.0
# HELP vllm:spec_decode_num_accepted_tokens_created Number of accepted tokens.
# TYPE vllm:spec_decode_num_accepted_tokens_created gauge
vllm:spec_decode_num_accepted_tokens_created{engine="0",model_name="qwen3.8-flash-next"} 1.789743398236043e+09
# HELP vllm:spec_decode_num_accepted_tokens_per_pos_total Accepted tokens per draft position.
# TYPE vllm:spec_decode_num_accepted_tokens_per_pos_total counter
vllm:spec_decode_num_accepted_tokens_per_pos_total{engine="0",model_name="qwen3.8-flash-next",position="0"} 10374.0
# HELP vllm:spec_decode_num_accepted_tokens_per_pos_created Accepted tokens per draft position.
# TYPE vllm:spec_decode_num_accepted_tokens_per_pos_created gauge
vllm:spec_decode_num_accepted_tokens_per_pos_created{engine="0",model_name="qwen3.8-flash-next",position="0"} 1.7897433982360573e+09