DeepSeek for the hard reasoning work.
Keep one OpenAI-compatible client and call DeepSeek on your own OpenRouter key. ProxyLLM wraps the traffic with spend tracking, budget caps, and request logs.
$129/month SaaS. Bring your own model keys. No inference markup.
Three steps to connect.
Add an OpenRouter key
DeepSeek ships through OpenRouter today. Paste your own OpenRouter key into ProxyLLM once; native DeepSeek key storage can land later without changing your client.
Use the OpenAI endpoint
Point any OpenAI-compatible client at https://api.proxyllm.ai/v1 and authenticate with your ProxyLLM key.
Pick the model per job
Call deepseek/deepseek-r1 for reasoning-heavy jobs and DeepSeek chat models for extraction, classification, and high-volume background work. The choice stays in your code.
One endpoint for DeepSeek.
The deepseek/ prefix passes through your configured OpenRouter key today.
from openai import OpenAI
client = OpenAI(
base_url="https://api.proxyllm.ai/v1",
api_key="pk_live_...", # your ProxyLLM key
)
r = client.chat.completions.create(
model="deepseek/deepseek-r1",
messages=[{"role": "user", "content": "Solve and explain this bug."}],
)Run your AI workloads on your ChatGPT subscription.
ProxyLLM runs OpenAI's Codex for you, signed in with your own ChatGPT account. Your apps call one OpenAI-compatible endpoint and the work bills to your flat plan instead of per-token API pricing.
Put DeepSeek behind a budget.
Set caps per sub-key and watch cost per request, with no markup on inference. OpenAI-bound work can run through Codex Hosted on your ChatGPT subscription; DeepSeek passes through on your own key.