Gahow Wang
445e491123
third_party/vllm/ now tracked in git for direct patch management.
Based on vLLM v0.18.1 release with one patch applied:
vllm/v1/core/sched/scheduler.py:
Replace fatal assert with graceful skip when KV transfer callback
arrives for an already-aborted request during PD disaggregated serving.
Future vLLM modifications should be made directly in third_party/vllm/
and committed normally. The patches/ directory is kept as documentation
of what changed from upstream.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
18 lines
637 B
Django/Jinja
18 lines
637 B
Django/Jinja
{%- set counter = namespace(index=1) -%}
|
||
{%- for message in messages -%}
|
||
{%- if message['role'] == 'user' -%}
|
||
{{- '[Round ' + counter.index|string + ']\n\n问：' + message['content'] -}}
|
||
{%- set counter.index = counter.index + 1 -%}
|
||
{%- endif -%}
|
||
{%- if message['role'] == 'assistant' -%}
|
||
{{- '\n\n答：' + message['content'] -}}
|
||
{%- if (loop.last and add_generation_prompt) or not loop.last -%}
|
||
{{- '\n\n' -}}
|
||
{%- endif -%}
|
||
{%- endif -%}
|
||
{%- endfor -%}
|
||
|
||
|
||
{%- if add_generation_prompt and messages[-1]['role'] != 'assistant' -%}
|
||
{{- '\n\n答：' -}}
|
||
{%- endif -%} |