ComfyUI Node
[llama.cpp] Create Runtime Session
Starts a workflow-owned local llama.cpp server and keeps its model loaded until Unload Session or prompt-end cleanup. The session owns its server API key.
[llama.cpp] Create Runtime Session
- model_profile
- hardware_profile
- reasoning
- speculative
- prefill_profile
- session
◄model_path[no GGUF models found]►
◄mmproj_path[none]►
◄n_ctx8192►
◄verbosefalse►
◄custom_chat_template—►
Categoryllama_cpp/session
Inputs (10)
| Name | Type | Default | Description |
|---|---|---|---|
| model_path | COMBO | [no GGUF models found] | 1 options: [no GGUF models found] |
| mmproj_path | COMBO | [none] | 1 options: [none] |
| n_ctx | INT | 8192512–1048576 | — |
| verbose | BOOLEAN | false | — |
| model_profileopt | OLLAMA_IMAGE_LIST_LLAMA_CPP_MODEL_PROFILE | — | |
| custom_chat_templateopt | STRING | Optional custom Jinja template for this server session. When connected, it overrides the GGUF metadata template; when disconnected, the GGUF template is used. | |
| hardware_profileopt | OLLAMA_IMAGE_LIST_LLAMA_CPP_HARDWARE_RUNTIME_PROFILE | — | |
| reasoningopt | OLLAMA_IMAGE_LIST_LLAMA_CPP_REASONING_CONFIG | — | |
| speculativeopt | OLLAMA_IMAGE_LIST_LLAMA_CPP_SPECULATIVE_CONFIG | — | |
| prefill_profileopt | OLLAMA_IMAGE_LIST_LLAMA_CPP_PREFILL_PROFILE | — |
Outputs (1)
| Name | Type | Description |
|---|---|---|
| session | OLLAMA_IMAGE_LIST_LLAMA_CPP_SESSION | — |