Quick Answer
DeepSeek V4 Pro speaks the OpenAI-compatible interface, so the integration pattern is one thing: point the base URL at the endpoint, fill in the key, pick the right model ID. With Moyi API:
Base URL : https://api.moyi-api.com/v1
API Key : sk-your-key
Model ID: deepseek-v4-pro # DeepSeek V4 Pro, paid
Any tool that supports a custom base URL — dsh, OpenCode, Cline, Cherry Studio, Codex — plugs in with this pattern, near-zero migration cost. Below: the endpoint, model IDs, and a per-tool quick reference.
New accounts start with a $0 balance — top up before your first request. Top up from $5. If a request returns 402 (400 on the OpenAI format) with "wallet balance is insufficient", your balance is empty — your base URL and key are fine. Top up and retry.
Why "One Endpoint, Every Tool"
The OpenAI-compatible interface is a de facto standard: most AI tools let you send requests to a custom base URL. That means no per-tool SDK or plugin — just:
- Set the tool's base URL to
https://api.moyi-api.com/v1; - Fill in your
sk-...key; - Pick a DeepSeek model ID.
Three steps and DeepSeek V4 Pro runs in that tool. That's why it landed in dsh, OpenCode, Cline, Cherry Studio, and Codex so quickly.
Model ID Quick Reference
| Model ID | Position | Details |
|---|---|---|
deepseek-v4-pro |
Flagship Pro, paid | Model page |
deepseek-v4-flash |
Cheap Flash, paid | Pricing page |
Note: DeepSeek's official price for V4 Pro is $1.32 per million input tokens and $3.96 per million output tokens at peak, half that off-peak (DeepSeek pricing, November 2, 2026). Moyi API's current prices are on the pricing page. For splitting work between the two models by difficulty, see V4 Pro vs Flash.
The Unified Pattern (Verify with curl)
First verify the endpoint with curl:
curl https://api.moyi-api.com/v1/chat/completions \
-H "Authorization: Bearer sk-your-key" \
-H "Content-Type: application/json" \
-d '{
"model": "deepseek-v4-pro",
"messages": [{"role": "user", "content": "Introduce yourself in one sentence"}]
}'
A standard OpenAI chat.completion response means endpoint, key, and model ID are all correct. On the Python side, reuse the OpenAI SDK:
from openai import OpenAI
client = OpenAI(
api_key="sk-your-key",
base_url="https://api.moyi-api.com/v1",
)
resp = client.chat.completions.create(
model="deepseek-v4-pro",
messages=[{"role": "user", "content": "Hello"}],
)
print(resp.choices[0].message.content)
Per-Tool Quick Reference
| Tool | Protocol | Config entry |
|---|---|---|
| DeepSeek Harness (dsh) | OpenAI | ~/.dsh/settings.yaml provider + ~/.dsh/.env |
| Codex | OpenAI | ~/.codex/config.toml (base_url) + OPENAI_API_KEY |
| OpenCode | OpenAI | opencode.json or env vars |
| Cline | OpenAI-compatible | VS Code extension → "OpenAI Compatible" |
| Cherry Studio | OpenAI-compatible | Add provider + base URL / key |
Per-tool steps: OpenCode / Cline / Cherry Studio, Codex, dsh.
FAQ
Does the base URL include /v1?
OpenAI SDK / tools: yes — https://api.moyi-api.com/v1. Anthropic-protocol tools: no — https://api.moyi-api.com. Match the protocol.
I get "Model 'deepseek-v4-pro-free' is not available". What now?
Moyi API has no deepseek-v4-pro-free model. Change the model ID to deepseek-v4-pro; your base URL and key stay the same.
Can one key power multiple tools?
Yes. One sk- key works in dsh, Codex, Cherry Studio, and more — one account, one bill.
Does the endpoint support streaming?
Yes. Add "stream": true for standard OpenAI SSE.
How do I debug 404 / 401?
404 → check the base URL has /v1; 401 → check the key starts with sk-, was copied fully, and wasn't deleted. Verify with the curl above first.
Register at Moyi API for a key — one endpoint, every tool.