Skip to content

Migration guide · 5 minutes

Fireworks AI → QuickSilver Pro

Both APIs are OpenAI-compatible, so the move is a base-URL swap. QuickSilver Pro serves the latest DeepSeek V4 wave — V4 Flash for low-cost chat and V4 Pro for premium reasoning — well below Fireworks' own-GPU rates. For the full side-by-side analysis, see /vs/fireworks.

Code changes
base_url · API key · model IDs
Model IDs to rename
1 mapping
Then
test your core flows end-to-end
  1. In your OpenAI SDK init, swap the base_url.

    diff
    - base_url="https://api.fireworks.ai/inference/v1"
    + base_url="https://api.quicksilverpro.io/v1"
  2. Fireworks uses fully-qualified account paths for model IDs. Use the structured mapping below to select the QuickSilver Pro short name.

    Fireworks AIQuickSilver Pro
    accounts/fireworks/models/deepseek-v4-prodeepseek-v4-pro

Full before/after

Before · Fireworks AI
import os
from openai import OpenAI

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

r = client.chat.completions.create(
    model="accounts/fireworks/models/deepseek-v4-pro",
    messages=[{"role": "user", "content": "Hi"}],
)
After · QuickSilver Pro
import os
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-v4-pro",
    messages=[{"role": "user", "content": "Hi"}],
)
import os
from openai import OpenAI

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

r = client.chat.completions.create(
    model="accounts/fireworks/models/deepseek-v4-pro",
    messages=[{"role": "user", "content": "Hi"}],
)

What you'll pay after switching

Per 1M tokens, input / output. QuickSilver Pro rates vs Fireworks AI's published per-token pricing.

ModelQuickSilver ProFireworks AISavings
DeepSeek V4 Pro$0.70 / $2.10$1.74 / $3.48~40% output

Fireworks AI prices checked 2026-08-12

Common migration pitfalls

  • FireFunction V2 has no equivalent

    FireFunction V2 is Fireworks' proprietary fine-tuned tool-calling model and is not in the QuickSilver Pro catalog. For tool calling, deepseek-v4.1-flash and qwen3.8-flash-next both support the standard OpenAI tools / function calling API.

  • Dedicated deployments and fine-tunes don't move

    QuickSilver Pro serves shared models only. Keep Fireworks dedicated deployments and fine-tuned models on Fireworks; the two SDKs coexist fine in one codebase.

  • Model generation may differ

    Fireworks and QuickSilver Pro both serve V4 Pro, but provider and quantization differences can affect output style. QuickSilver Pro also offers the official V4 Flash 0731 agent build. Re-run your evals after switching.

  • Rate limits work differently

    QuickSilver Pro applies per-key throughput caps (default 600 req/min, 1M tok/min, 8–100 parallel per account by top-up tier). For bursty traffic, enable retry-on-429 and request a higher limit if needed.

Migrating from Fireworks AI — FAQ

Start saving in 5 minutes

First top-up matched 100%, up to $50 in bonus credits, added on your next top-up. Keep your code on the OpenAI SDK — only the base URL and key change.