Anima LM Studio (prompt LLM)

Sends your idea to a local LLM server (LM Studio) and returns a structured answer: rating, count, character, tags by group and a short description. Connect it to Anima Prompt Builder.

By GUntilB·Created 2 days ago·Updated a day ago· 12

llm_text
status

◄idea►

◄system_prompt►

◄model▾►

◄temperature0.50►

◄unload_aftertrue►

◄use_cachetrue►

◄seed0►

◄server_urlhttp://127.0.0.1:1234►

◄max_tokens1024►

◄context_length8192►

◄model_override►

◄api_key►

◄prompt_profilesfw►

Categoryanima

Inputs (13)

Name	Type	Default	Description
idea	STRING		What should be in the picture, in any language.
system_prompt	STRING		Your own instructions for the LLM. Used only when prompt_profile is 'custom'; sfw and nsfw take their text from the prompts folder and ignore this field.
model	COMBO		Model on the LLM server. The list is read when the node loads; press R to refresh it.
temperature	FLOAT	0.500–2	Randomness of the answer. Lower is steadier; 0.5 works well for tag lists.
unload_after	BOOLEAN	true	Unload the model after the answer to free VRAM for image generation.
use_cache	BOOLEAN	true	Reuse the previous answer when the request and all settings are unchanged, without calling the LLM.
seed	INT	00–2147483647	Seed sent to the server. The same seed and settings give the same answer; it is also part of the cache key.
server_url	STRING	http://127.0.0.1:1234	Address of the LM Studio server.
max_tokensopt	INT	1024128–16384	Answer length limit. Raise it if the status says the answer was cut off.
context_lengthopt	INT	81922048–131072	Context size used when this node loads the model itself.
model_overrideopt	STRING		Type a model name here to use it instead of the one chosen in the list.
api_keyopt	STRING		API key, only if the server requires one.
prompt_profileopt	COMBO	sfw	Instruction set for the LLM: sfw (clothed, no explicit content), nsfw (all ratings), custom (uses the system_prompt field).

Outputs (2)

Name	Type	Description
llm_text	STRING	—
status	STRING	—