OpenRouter API (Tlant V1)

Call OpenRouter API with automatic caching. Same inputs + same seed = skip API call.

By Tlant·Created 2 years ago·Updated 4 months ago· 2

◄model_namexiaomi/mimo-v2-flash►

◄base_urlhttps://openrouter.ai/api/v1/chat/completions►

◄api_key►

◄system_promptYou are a helpful assistant.►

◄user_prompt►

◄temperature0.70►

◄max_tokens2048►

◄seed0►

◄remove_think_tagsfalse►

◄proxy_url►

◄top_p1.00►

◄timeout300►

◄image_detailauto►

◄user_prompt_input—►

◄system_prompt_input—►

CategoryTlant/API

Inputs (16)

Name	Type	Default	Description
model_name	STRING	xiaomi/mimo-v2-flash	OpenRouter model identifier, e.g. google/gemini-2.5-pro-preview
base_url	STRING	https://openrouter.ai/api/v1/chat/completions	API endpoint URL
api_key	STRING		OpenRouter API key (sk-or-v1-...)
system_prompt	STRING	You are a helpful assistant.	System prompt to set model behavior
user_prompt	STRING		User prompt / instruction
temperature	FLOAT	0.700–2	Controls randomness. Lower = more deterministic
max_tokens	INT	20481–128000	Maximum number of tokens in the response
seed	INT	00–18446744073709550000	Change seed to force re-execution; keep fixed to use cache
remove_think_tags	BOOLEAN	false	Remove <think>...</think> blocks from output
proxy_url	STRING		Optional proxy. e.g. http://127.0.0.1:1080 or socks5://127.0.0.1:1080
top_p	FLOAT	1.000–1	Nucleus sampling threshold
timeout	INT	30010–1200	Request timeout in seconds
image_detail	COMBO	auto	Image detail level for vision models
imagesopt	IMAGE		Optional image input for vision models. Supports batch.
user_prompt_inputopt	STRING		External user prompt input, will be appended to user_prompt
system_prompt_inputopt	STRING		External system prompt input, will be appended to system_prompt

Name	Type	Description
response_text	STRING	The model's text response
reasoning_content	STRING	Reasoning content returned by reasoning models (if available)