DeepSeek-V3.2 is the official successor to V3.2-Exp, designed for balanced inference and reasoning capabilities, achieving performance comparable to GPT-5.
Endpoint: POST https://api.skillboss.co/v1/run · Model: deepseek/deepseek-v3.2
Use cURL when you want the clearest raw HTTP example. This is the fastest way to validate auth, inputs, output shape, and the canonical endpoint before wiring the API into code or agent workflows.
curl -X POST https://api.skillboss.co/v1/run \
-H "Authorization: Bearer YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "deepseek/deepseek-v3.2",
"inputs": {
// Your parameters here
}
}'Start using deepseek/deepseek-v3.2 in under 60 seconds
FOR AI AGENTS
Base URL: https://api.skillboss.co/v1
Model: deepseek/deepseek-v3.2
Auth: Bearer token — get at skillboss.co/console
Pricing: $0.00/$0.00 per 1M tokens, no subscription
Or auto-configure everything:
set up skillboss.co/skill.mdFOR DEVELOPERS
Get your API key at skillboss.co/console
POST to api.skillboss.co/v1 with model deepseek/deepseek-v3.2
Parse the JSON response. Auth, billing, and rate limits handled automatically.
Power assistants, copilots, and agent loops
Generate text, summaries, and reasoning outputs
Route prompts through one unified API key
deepseek/deepseek-v3.2 is a chat api available through SkillBoss. Teams use it to add deepseek/deepseek-v3.2 to apps, automations, and AI agent workflows without managing separate vendor credentials. This page gives both humans and agents one stable place to understand the capability, pricing path, and invocation pattern before they ship it into production.
Send a POST request to https://api.skillboss.co/v1/run with "model": "deepseek/deepseek-v3.2" and an "inputs" object. SkillBoss handles auth, billing, and the unified API layer. If you are testing the endpoint manually, start with the cURL or Python tab below. If you are wiring an autonomous agent, use the x402 or MPP tabs when the payment path matters.
SkillBoss gives you one API key, pay-as-you-go billing, and a consistent integration flow across many APIs. That reduces setup time and makes the endpoint easier to use in Claude Code, Cursor, Windsurf, and other agents. It also gives you a canonical public page, consistent docs, and one billing surface instead of forcing each team to juggle provider-specific auth and pricing rules.
Yes. This page is designed as a discovery surface for both humans and agents: clear capability framing, direct code examples, and a stable canonical URL make it easier to find, evaluate, and invoke. The structure is deliberate so search engines, agent runtimes, and developers can all understand what the endpoint does, how it is paid for, and which next pages matter.
deepseek/deepseek-v3.2 is better served through the standard SkillBoss API key flow. x402 and MPP work best when pricing is deterministic before execution, which is not the safest assumption for this endpoint.
One API key, 100+ models, zero markup
AI Agents
Tell your agent:
set up skillboss.co/skill.mdAuto-configures base URL, auth, and deepseek/deepseek-v3.2 access. Works with Claude Code, Cursor, Windsurf, and any MCP-compatible agent.
Developers
1. Get API key at skillboss.co/console
2. POST to api.skillboss.co/v1/run
3. Model: deepseek/deepseek-v3.2
Pay-as-you-go. $0.00/$0.00 per 1M tokens. No subscription required.