Original vllm ops.reshape_and_cache passes 8 args: key, value, key_cache, value_cache, slot_mapping, kv_cache_dtype, k_scale, v_scale Our patch had 7 (single kv_scale) → TypeError on first request → engine dead.
Original vllm ops.reshape_and_cache passes 8 args: key, value, key_cache, value_cache, slot_mapping, kv_cache_dtype, k_scale, v_scale Our patch had 7 (single kv_scale) → TypeError on first request → engine dead.