DeepSeek Chat
General chat, coding help, lightweight agent tasks, and API migration from OpenAI-style SDKs.
$0.30 input / 1M tokens
$0.60 output / 1M tokens
DeepSeek API for global builders
Chinese models, OpenAI-compatible.
Start with DeepSeek Chat through a familiar /v1 API, pay with prepaid USD credits, create API keys in the customer console, and track usage without juggling Chinese provider accounts.
Model store
Start with the routes customers can understand.
A focused catalog beats a noisy marketplace during launch. Each model card shows status, pricing, and the action that matters: create a key in the console.
DeepSeek Reasoner
Reasoning-heavy workloads, multi-step analysis, tool planning, and harder coding tasks after route verification.
Pricing pending
Published after provider and billing checks
Qwen, Kimi, GLM, Doubao
Additional Chinese model routes will be added only when pricing, fallback behavior, and billing records are ready.
Roadmap
Model cards will be added one route at a time
How it works
One clean customer path, no test dashboard detours.
The public website should sell and explain. The console should handle identity, credits, keys, and usage. That keeps the product path legible.
2
Add credits
Recharge prepaid balance before production traffic.
3
Create API key
Generate a token and keep it secret in your app environment.
4
Call DeepSeek
Use the OpenAI SDK against https://api.modelbridgeapi.com/v1.
Positioning
Not an account pool. A focused bridge to Chinese models.
ModelBridge should win a narrower job than all-model routers: make Chinese model access clear, payable, and usable for global developers.
No account sharing
Use official or compliant upstream APIs. Do not rely on shared subscriptions, browser reverse engineering, or hidden account pools.
DeepSeek first
Start with the most concrete route, then add more Chinese providers only after verification.
Usage ledger
Track requests, tokens, charged amount, latency, and route behavior in the customer console.
Prepaid credits
Keep billing exposure explicit for developers testing Chinese model routes.
Quickstart
Switch an OpenAI client in three lines.
Create a token in the console, set the hosted base URL, and call deepseek-chat.
from openai import OpenAI
client = OpenAI(
api_key="YOUR_MODELBRIDGE_API_KEY",
base_url="https://api.modelbridgeapi.com/v1"
)
response = client.chat.completions.create(
model="deepseek-chat",
messages=[{"role": "user", "content": "Hello"}]
)
print(response.choices[0].message.content)