lukaszraczylo/llm-testbench
1
0
Fork 0
mirror of https://github.com/lukaszraczylo/llm-testbench.git synced 2026-10-14 00:57:45 +00:00
Code Issues Packages Projects Releases Wiki Activity

fix(runner): record wall-clock latency on error rows

Browse Source 
An error row with zero latency reads as an instant rejection when it
can be three exhausted request timeouts. A 651-second thinking-model
generation looked instant in the 2026-08-30 run and misdirected the
diagnosis toward gateway rejection. The error path now records the
wall clock across all attempts.
This commit is contained in:
lukaszraczylo
2026-08-30 16:19:21 +01:00
parent 7616f4bdee
commit 6ad9dfbe42
2 changed files with 7 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files
config.example.yaml
+1 -1
View File
@@ -5,7 +5,7 @@ models:
- uni/qwen3.6-27b
- uni/btl-4
concurrency: 8 # default when omitted; matches internal/config's default
request_timeout: 300s # default when omitted
request_timeout: 900s # default when omitted
max_tokens_default: 12000 # default when omitted; a reasoning model can spend
# thousands of tokens on reasoning_content before
# writing an answer, so this must stay generous