Home / Migrate / From OpenRouter
Migration guide · 5 minutes

OpenRouter → QuickSilver Pro

Two lines of code. ~20% cheaper on DeepSeek V3, DeepSeek R1, and Qwen3.5-35B-A3B. This is the action guide — for the comparative analysis, see /vs/openrouter.

The 5 steps

  1. 1

    Get a QuickSilver Pro API key

    Sign up at quicksilverpro.io/dashboard. New accounts get $1 in free credits — enough for ~500-700 real DeepSeek V3 calls to evaluate before paying. Minimum top-up is $5.

  2. 2

    Change the base URL

    In your OpenAI SDK init, swap the base_url.

    - base_url="https://openrouter.ai/api/v1"
    + base_url="https://api.quicksilverpro.io/v1"
  3. 3

    Swap the API key

    Replace OPENROUTER_KEY with your QSP key (from step 1).

    - api_key=os.environ["OPENROUTER_KEY"],
    + api_key=os.environ["QSP_KEY"],
  4. 4

    Rename model IDs

    Drop the provider/ prefix.

    OpenRouter QuickSilver Pro
    deepseek/deepseek-chatdeepseek-v3
    deepseek/deepseek-r1deepseek-r1
    qwen/qwen3.5-35b-a3bqwen3.5-35b
  5. 5

    Test your core flows end-to-end

    Run one representative request for each feature your app uses — chat, tool_calls, json_schema strict mode, streaming. All standard OpenAI features work on QSP; any behavioral diff is a bug — report it to us and we'll fix it.

Full before/after

Before · OpenRouter
from openai import OpenAI

client = OpenAI(
    base_url="https://openrouter.ai/api/v1",
    api_key=os.environ["OPENROUTER_KEY"],
)

r = client.chat.completions.create(
    model="deepseek/deepseek-chat",
    messages=[{"role": "user", "content": "Hi"}],
)
After · QuickSilver Pro
from openai import OpenAI

client = OpenAI(
    base_url="https://api.quicksilverpro.io/v1",
    api_key=os.environ["QSP_KEY"],
)

r = client.chat.completions.create(
    model="deepseek-v3",
    messages=[{"role": "user", "content": "Hi"}],
)

Common migration pitfalls

Cache-hit ratio drops temporarily
OpenRouter's cache lives across upstream providers; when you switch to QSP your cache starts cold. For the first day or two, observed cost may be a few % higher than list-price math suggests. Cache fills within 24-48h of your normal traffic pattern.
Model version drift
OpenRouter's deepseek/deepseek-chat occasionally points to different DeepSeek V3 revisions over time. QSP routes deepseek-v3 to a specific revision. If your evals were tuned to a particular DeepSeek output style, re-run them after switching.
Don't migrate models you don't use on QSP
QSP only serves DeepSeek V3, DeepSeek R1, and Qwen3.5-35B-A3B. If your OpenRouter setup also uses GPT-4, Claude, Llama, or Mistral — keep calling OpenRouter for those. Many teams run both SDKs side-by-side and route per-request.
Rate limits work differently
OpenRouter silently smooths across upstreams. QSP has its own per-key rate limit (default 600 req/min). For bursty workloads, enable retry-on-429 in your client and monitor 429 rate; ask for a higher limit if needed.

Need help?

Email hello@quicksilverpro.io — a human replies usually within 4 hours. For the broader comparative analysis (when to migrate, when to stay), see QuickSilver Pro vs OpenRouter.

Start saving in 5 minutes

$1 free to try, no card required. Keep your OpenAI SDK unchanged — only the URL and key change.

Get API Key