fix: align Dockerfile + yaml with wudixzy/competition upstream

Dockerfile:
- Add ENV: PATH, PYTHONPATH, LD_LIBRARY_PATH (corex SDK discovery)
- Add ENV: ENABLE_CUSTOM_IPC=1 (TP inter-process communication)
- Add ENV: BI100_PREFIX_* (prefix caching fingerprint)
- Add ENV: PYTHONUNBUFFERED=1, PYTHONFAULTHANDLER=1
- Change RUN to: cd ./qwen3_6_scripts && bash ./patch_ops.sh (match wudixzy)

computility-run.yaml:
- max-num-seqs: 2 → 1 (wudixzy upstream value)
  n=2 is handled by serving_chat.py _sequential_greedy_fanout
  which runs two n=1 requests and merges. Requires max_num_seqs=1.
  max_num_seqs=2 bypassed the fanout → vllm rejected greedy n=2 → HTTP 400

patch_ops.sh:
- set -eo → set -euo (match wudixzy)
This commit is contained in:
project6-dev
2026-08-12 04:26:09 +00:00
parent d025b08a95
commit a33060bc5e
3 changed files with 10 additions and 7 deletions

View File

@@ -28,7 +28,7 @@
# --max-seq-len-to-capture 32768 --enable-auto-tool-choice \
# --tool-call-parser qwen3_coder --reasoning-parser qwen3
set -eo pipefail
set -euo pipefail
# cd into this script's directory so ./relative paths work
cd "$(dirname "${BASH_SOURCE[0]}")"