From a3c1342837a5d50bf738041d37d138cce8b34147 Mon Sep 17 00:00:00 2001 From: Hermes Agent service account Date: Thu, 6 Aug 2026 09:38:32 -0500 Subject: [PATCH] llm-inference-multimodel: reset qwen shadow unit to disabled by default -- model disqualified (n_ctx_train=32768, not 64K+), leaving enabled would crash-loop on next playbook run --- ansible/roles/llm-inference-multimodel/defaults/main.yml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/ansible/roles/llm-inference-multimodel/defaults/main.yml b/ansible/roles/llm-inference-multimodel/defaults/main.yml index 74ef9e8..377659c 100644 --- a/ansible/roles/llm-inference-multimodel/defaults/main.yml +++ b/ansible/roles/llm-inference-multimodel/defaults/main.yml @@ -71,7 +71,7 @@ llm_toolcall_model_id: mistral-small-24b-instruct-2501 # (either free VRAM by stopping llama-server-toolcall for the shadow window, # or reduce ctx-size). llm_qwen_service_enabled defaults to false so a normal # playbook run will template the unit but NOT start it. -llm_qwen_service_enabled: true +llm_qwen_service_enabled: false llm_qwen_port: 8002 llm_qwen_model_path: "{{ llm_models_dir }}/Qwen2.5-14B-Instruct-Q5_K_M.gguf" llm_qwen_model_url: "https://huggingface.co/bartowski/Qwen2.5-14B-Instruct-GGUF/resolve/main/Qwen2.5-14B-Instruct-Q5_K_M.gguf"