Cloudflare decision route
clef-flash
Clef Flash API pricing.
Cloudflare's 9B decision model for latency-critical, hot-path decisions: route, block, escalate. Cloudflare measured it 13x faster than Jev at the median. Same request shape as Jev, billed on input only.
Input / 1M tokens · output free
Input / 1M tokens
$0.038Free
Output / 1M tokens
Free
Context window
24K
Provider
Cloudflare
Question types
Ask in types, branch in code.
Every question is evaluated in parallel and in isolation against the same state, so adding questions barely changes latency. Mix all three types in one call.
Billed on input only
Output tokens free
Same key as chat models
POST /v1/systemone
First request
One call, three decisions.
Name each question; its answer comes back under the same id. Choice and score answers carry a confidence your code can gate on.
curl https://api.routerplex.com/v1/systemone \-H "Authorization: Bearer $ROUTERPLEX_API_KEY" \-H "Content-Type: application/json" \-d '{"model": "clef-flash","state": "Help! My payouts have been failing for 3 days.","questions": {"is_urgent": {"type": "noul", "instructions": "Does this convey urgency?"},"team": {"type": "choice","instructions": "Which team should handle this?","criteria": {"billing": "Payments, refunds", "technical": "Bugs, outages", "sales": "Pricing"}},"frustration": {"type": "score","instructions": "How frustrated is the customer?","criteria": ["Calm", "Frustrated", "Very angry"]}}}'
{"model": "clef-flash","answers": {"is_urgent": {"type": "noul", "noul": 0.95},"team": {"type": "choice", "choice": "billing", "confidence": 0.96,"probabilities": {"billing": 0.98, "technical": 0.02, "sales": 0.0}},"frustration": {"type": "score", "score": 1.04, "confidence": 0.95,"legend": {"0": "Calm", "1": "Frustrated", "2": "Very angry"},"probabilities": {"0": 0.0, "1": 0.96, "2": 0.04}}},"usage": {"input_tokens": 382, "output_tokens": 73, "cost": 0.0}}
Route notes
Questions before production traffic.
01How much does Clef Flash cost on RouterPlex?+
Nothing during the launch promotion. The list price is $0.038 per 1M input tokens with output free, and that is what RouterPlex will charge when the promotion ends. Calls still need an account with an active balance.
02Can I call it through /v1/chat/completions?+
No. Decision models do not generate text. Send a state and typed questions to POST /v1/systemone on the same base URL and the same API key you use for chat models.
03Does the TypeSafe SDK work with RouterPlex?+
The request and response shapes match TypeSafe's System One API, so a TypeSafe client works once you set its base URL (or TYPESAFE_BASE_URL) to https://api.routerplex.com and pass your RouterPlex key.
Decision models
Other decision routes.
Deploy this route
Put Clef Flash behind one key.
Top up from $5 with no subscription required.