When to use it
More demanding reasoning and longer-running agent work. Validate quality against your own repository or production traces before scaling.
DeepSeek route
DeepSeek V4 Pro
DeepSeek V4 Pro through an OpenAI-compatible Zoryntel endpoint, with named upstream policy and transparent current pay-as-you-go pricing.
Available
DeepSeek
DeepSeek V4 Pro
More demanding reasoning and longer-running agent work.
- Model ID
deepseek-v4-pro- API format
- OpenAI-compatible
- Upstream policy
- Named route; no silent substitution
Input$0.348USD / 1M · list $0.435
Output$0.696USD / 1M · list $0.870
Cache hit$0.00290USD / 1M · list $0.003625
Current pay-as-you-go rate: 20% below the verified provider list-price snapshot. View sources and change history.
What is disclosed
The public model ID, intended upstream, cache rate and substitution policy are documented. A material route or model change requires an explicit version update.
Beta limitation
Availability, throughput and compatibility are still being tested. We do not publish unmeasured SLA or benchmark claims.
cURL
curl https://api.funly-ai.com/v1/chat/completions \
-H "Authorization: Bearer $ZORYNTEL_API_KEY" \
-H "Content-Type: application/json" \
-d '{"model":"deepseek-v4-pro","messages":[{"role":"user","content":"Hello"}]}'