chat_completion
Send chat completion requests to DeepSeek V4 Flash via proxy. Provide messages in OpenAI format to receive AI responses with token usage and cost estimates.
Instructions
Send a chat completion request to DeepSeek V4 Flash through the GadgetHumans proxy. REQUIRES an API key with credits. Get one via wallet_create or wallet_buy.
Use this to get AI responses. Each call costs ~1 credit from your wallet.
Accepts an OpenAI-compatible messages format. Returns response with
content, token usage, and cost estimate.
Parameters:
messages — JSON string of messages array (e.g. '[{"role":"user","content":"Hello"}]').
Must be valid JSON with at least one message.
model — Model to use (default: "deepseek-v4-flash").
temperature — Sampling temperature (0.0-1.0, default: 0.7).
max_tokens — Maximum tokens in response (default: 2000).
api_key — Your wallet API key (starts with 'gh_'). REQUIRED for paid access.
Input Schema
TableJSON Schema
| Name | Required | Description | Default |
|---|---|---|---|
| messages | Yes | ||
| model | No | deepseek-v4-flash | |
| temperature | No | ||
| max_tokens | No | ||
| api_key | No |
Output Schema
TableJSON Schema
| Name | Required | Description | Default |
|---|---|---|---|
| result | Yes |