Integrations

Use QuickSilver Pro from your favorite tool

Anything that speaks the OpenAI Chat Completions API works with QSP. DeepSeek V4 Flash also supports the Responses API used by Codex.

The two values you always need

Every tool below just wants these two strings — the rest of the config is the tool's own UI:

shell

Base URL  →  https://api.quicksilverpro.io/v1
API key   →  sk-...   (from your dashboard)

Don't see your tool below? Anything OpenAI-compatible works.

Codex (Responses API)

DeepSeek V4 Flash is available through the Responses API used by Codex. Add a QSP provider to ~/.codex/config.toml:

toml

model = "deepseek-v4-flash"
model_provider = "quicksilverpro"
preferred_auth_method = "apikey"
forced_login_method = "api"
model_reasoning_effort = "high"

[model_providers.quicksilverpro]
name = "QuickSilver Pro"
base_url = "https://api.quicksilverpro.io/v1"
wire_api = "responses"
experimental_bearer_token = "<your QSP API key>"

Codex sends requests to /v1/responses; the same QSP key and model ID also work through Chat Completions.

Aider (terminal pair-programmer)

Aider speaks OpenAI natively. Point it at QSP with two env vars and a model flag — that's it.

shell

export OPENAI_API_BASE="https://api.quicksilverpro.io/v1"
export OPENAI_API_KEY="$QSP_KEY"

aider --model openai/deepseek-v4-flash

The openai/ prefix tells Aider to route through the OpenAI client; everything after it is the QSP model ID (see the list).

Cline / Roo Code (VS Code agent)

Open the Cline (or Roo Code) sidebar → settings cog → API Provider:

text

API Provider:  OpenAI Compatible
Base URL:      https://api.quicksilverpro.io/v1
API Key:       <your QSP key>
Model ID:      deepseek-v4-flash   (or any QSP model)

Reasoning-by-default models (DeepSeek V4 Flash, DeepSeek V4 Pro, Qwen3.8 Max, Qwen3.7 Max, Qwen3.7 Plus) work as Cline's "Thinking" provider out of the box.

Continue.dev (VS Code / JetBrains)

Add QSP as a model in ~/.continue/config.json:

json

{
  "models": [
    {
      "title": "DeepSeek V4 Pro (QSP)",
      "provider": "openai",
      "model": "deepseek-v4-flash",
      "apiBase": "https://api.quicksilverpro.io/v1",
      "apiKey": "$QSP_KEY"
    }
  ]
}

Open WebUI

In Open WebUI: Admin Panel → Connections → OpenAI API, add a new connection:

text

URL:  https://api.quicksilverpro.io/v1
Key:  <your QSP key>

Save and reload — all 40 QSP models will appear in the model picker.

LobeChat

Settings → Language Model → enable OpenAI provider, then override:

text

API Proxy URL:  https://api.quicksilverpro.io/v1
API Key:        <your QSP key>
Model List:     deepseek-v4-flash,deepseek-v4-pro,qwen3.8-max,qwen3.7-max,qwen3.7-plus,qwen3.7-flash,qwen3.6-plus,qwen3.6-35b,kimi-k2.6,kimi-k2.7-code,kimi-k3,muse-spark-1.2,muse-glimmer-30b,glm-5.3,glm-5.2,nemotron-3.5-lightning,gpt-5.6-luna,gpt-5.6-terra,gpt-5.6-sol,grok-4.5,grok-4.6,minimax-m3,mimo-v2.5,hy3,claude-opus-5,claude-fable-5,claude-opus-4-8,claude-opus-4-6,claude-sonnet-4-6,claude-sonnet-5,claude-haiku-4-5,gemini-3.7-flash,gemini-3.6-flash,gemini-3.5-flash-lite,gemini-3.5-flash,gemini-3.1-pro-preview,gemini-3-pro-image,gemini-3-flash-preview,gemini-3.1-flash-lite,flux.2-pro,
                gemini-3.1-pro-preview,gemini-3.5-flash,
                gemini-3.5-flash-lite,gemini-3-pro-image

Anything-LLM

Workspace → LLM Preference → Generic OpenAI:

text

Base URL:  https://api.quicksilverpro.io/v1
API Key:   <your QSP key>
Model:     deepseek-v4-flash   (or any QSP model)

Cursor (custom OpenAI endpoint)

Cursor Settings → Models → enable Override OpenAI Base URL:

text

Override OpenAI Base URL:  https://api.quicksilverpro.io/v1
OpenAI API Key:            <your QSP key>

Add the QSP model ID under Add custom model — Cursor will route through your override base URL for any custom model.

LangChain / LlamaIndex / Vercel AI SDK

Pass QSP's base URL into the OpenAI provider for any of these frameworks. Example (LangChain Python):

python

import os
from langchain_openai import ChatOpenAI

llm = ChatOpenAI(
    model="deepseek-v4-flash",
    base_url="https://api.quicksilverpro.io/v1",
    api_key=os.environ["QSP_KEY"],
)

Anything else

QSP is OpenAI-compatible — Chat Completions, streaming, tool calling, structured outputs, and usage accounting across the catalog, plus the Responses API for DeepSeek V4 Flash. If your tool talks to api.openai.com/v1, changing base_url to https://api.quicksilverpro.io/v1 is almost always all you need.

Hit a snag? hello@quicksilverpro.io — we'll either help you wire it or fix the gateway.