ccapi-ai DeepSeek LLM Integration API
Call the official DeepSeek V4 API through CCAPI with one token. --- ## Authentication Use `Authorization: Bearer `. --- ## Recommended endpoint - `POST /v1/chat/completions` --- ## Current models - `deepseek-v4-flash`: high-throughput, low-latency default - `deepseek-v4-pro`: complex reasoning and agentic coding Both models provide a 1M-token context window, thinking and non-thinking modes, JSON output, tool calls, SSE streaming, and cache-aware usage fields. Thinking is enabled by default. The retired `deepseek-chat` and `deepseek-reasoner` identifiers are not used in new integrations. See `GET /v1/models` for live availability. --- ## Pricing Visit the [Pricing page](/pricing?vendor=DeepSeek) to see full model pricing for this provider.