chat_completion
Chat Completion
Generate a conversational response using DeepSeek V4 models. Supports non-thinking chat, thinking-mode reasoning, JSON output, and function/tool calling.
Ask your agent in plain language. Each of these maps to a real tool on the server.
chat_completion
Generate a conversational response using DeepSeek V4 models. Supports non-thinking chat, thinking-mode reasoning, JSON output, and function/tool calling.
fim_completion
Fill In the Middle (FIM) completion for code and content. Provide a prefix and optional suffix, and DeepSeek generates the content that belongs in between.
get_balance
Retrieve the current DeepSeek account balance and availability status. Returns balance details broken down by currency, including granted and topped-up amounts.
list_models
List all currently available DeepSeek models with their metadata. Returns model identifiers that can be used in chat and FIM completion requests.
chat_prefix_completion
Continue from a supplied assistant prefix using DeepSeek beta chat prefix completion. Useful for forcing an answer to start with a code fence, template, or partial response.
Every client uses the same Magic MCP URL, and so does any other MCP client. Add it once and sign in with SSO.
{"mcpServers": {"metorial": {"url": "<your-magic-mcp-url>"}}}
claude mcp add --transport http metorial <your-magic-mcp-url>
codex mcp add metorial --url <your-magic-mcp-url>codex mcp login metorial
One governed layer for every app your agents reach, not a separate connector per tool.
Generate non-streaming chat completions, thinking-mode reasoning responses, and code completions using DeepSeek V4 models. Supports structured JSON output, function calling with up to 128 tool definitions, beta strict tool-call mode, beta FIM completion, and beta chat prefix completion. List available models and check account balance. Compatible with the OpenAI-format DeepSeek API.
Common questions about connecting Deepseek to AI agents with Metorial.