Entrpi 3f05a87cdf fix(serve): let empty TOOL_PARSER/REASONING_PARSER disable the parser ...

Use ${VAR-default} instead of ${VAR:-default} so an explicit empty value
(e.g. REASONING_PARSER=) disables the parser rather than falling back to the
default — matching the README's documented "disable with TOOL_PARSER=''".

2026-06-27 17:29:27 +10:00

..

mtp_serve.sh

serve: enable OpenAI auto tool-calling (qwen3_xml) + reasoning split by default

2026-06-24 23:08:31 +10:00

patch_fla_shmem.py

qwen3.5-122B-A10B on DGX Spark: vLLM + DFlash + dense-bandwidth stack, one-shot installer

2026-06-24 13:02:35 +10:00

patch_inc_hybrid.py

qwen3.5-122B-A10B on DGX Spark: vLLM + DFlash + dense-bandwidth stack, one-shot installer

2026-06-24 13:02:35 +10:00

patch_int8_lmhead_v3.py

kv: recover dense pool 376k->427k (+13%) at the same 0.82 headroom

2026-06-24 17:41:07 +10:00

patch_unify2.py

qwen3.5-122B-A10B on DGX Spark: vLLM + DFlash + dense-bandwidth stack, one-shot installer

2026-06-24 13:02:35 +10:00

serve.sh

fix(serve): let empty TOOL_PARSER/REASONING_PARSER disable the parser

2026-06-27 17:29:27 +10:00