thinkneo route model

AI Smart Router — find the cheapest model that meets your quality threshold. Specify your task type and quality requirements, and ThinkNEO will recommend the op

liveunknown~5s typical

mcpcontrol-plane

$0.0020

per call

plus $0.001 platform fee per run

Open MCP server; Cracked bills routing only

Provider Upstream docs

$Set up https://cracked-ai-pearl.vercel.app/SKILL.md, then use Cracked to run mcp-ai-thinkneo-control-plane /thinkneo_route_model for me.

Input

Field	Type	Description
task_type*	string	The type of AI task: summarization, classification, code_generation, chat, analysis, trans
quality_threshold	integer	Minimum quality score required (0-100). Default 85 = enterprise-grade. default 85
max_latency_ms	string	Maximum acceptable latency in milliseconds. Omit for no limit.
budget_per_request	string	Maximum budget per request in USD. Omit for no limit.
preferred_providers	string	Comma-separated list of preferred providers (e.g., 'openai,anthropic'). These will be prio
estimated_tokens	integer	Estimated total tokens for the request (input + output). Default 1000. default 1000
text_sample	string	Optional sample text for better routing. Helps estimate token count and task complexity. M

Call it

curl

curl https://cracked-ai-pearl.vercel.app/v1/run \
  -H "Authorization: Bearer ck_live_..." -H "content-type: application/json" \
  -d '{"provider":"mcp-ai-thinkneo-control-plane","endpoint":"/thinkneo_route_model","input":{"task_type":"...","quality_threshold":85,"estimated_tokens":1000}}'

cli

npx cracked-ai run -p mcp-ai-thinkneo-control-plane -e /thinkneo_route_model -i '{"task_type":"...","quality_threshold":85,"estimated_tokens":1000}'

mcp

run_tool({ provider: "mcp-ai-thinkneo-control-plane", endpoint: "/thinkneo_route_model", input: {"task_type":"...","quality_threshold":85,"estimated_tokens":1000} })

Try it

Runs against your signed-in workspace balance. Sign in if you have not.