Launch bonus

Open-source inference,20% below the rest.

The 9 most popular open-source models — DeepSeek V4 Flash & Pro, V3, R1, Qwen 3.7 Max + 3.6 Plus + 3.6 + 3.5, Kimi K2.6 — through an OpenAI-compatible API. Cheaper than every other reseller. Change one line of code.

Get API KeyView Pricing

or try the models live on HuggingFace — no signup required.

  • No subscription
  • OpenAI compatible
  • Pay as you go
  • Text + images
Drop-in with
  • OpenAI SDK
  • Aider
  • Cursor
  • Cline
  • Continue.dev
  • LangChain
  • Vercel AI SDK
python
1# One line change. That's it.
2from openai import OpenAI
3 
4client = OpenAI(
5    base_url="https://api.quicksilverpro.io/v1",
6    api_key="your-api-key",
7)
Pricing

Cheapest open-source inference

Per 1M tokens for text models · per image or per audio minute where noted.

Model
Context
Input
Output
Savings
DeepSeek V4 FlashNew
deepseek-v4-flash
fast chat & coding, 1M context, thinking on by default
1M
$0.08$0.10
$0.16$0.20
−20%
DeepSeek V4 ProNew
deepseek-v4-pro
premium reasoning, 1M context
1M
$0.348$0.435
$0.696$0.87
−20%
DeepSeek V3
deepseek-v3
chat, coding, structured output
128K
$0.16$0.20
$0.616$0.77
−20%
DeepSeek R1Reasoning
deepseek-r1
math, multi-step reasoning, logic
128K
$0.56$0.70
$2.00$2.50
−20%
Qwen3.7 MaxNew
qwen3.7-max
Qwen 3.7 flagship, agent / coding, 1M context
1M
$2.00$2.50
$6.00$7.50
−20%
Qwen3.6 PlusNew
qwen3.6-plus
thinks-by-default flagship, 1M context
1M
$0.26$0.325
$1.56$1.95
−20%
Qwen3.6-35B-A3B
qwen3.6-35b
long-context RAG, drop-in 3.5 upgrade
262K
$0.12$0.15
$0.80$1.00
−20%
Qwen3.5-35B-A3B
qwen3.5-35b
long-context RAG, summarization
262K
$0.111$0.139
$0.80$1.00
−20%
Kimi K2.6
kimi-k2.6
Opus-class agentic / planning
256K
$0.584$0.73
$2.79$3.49
−20%
Claude Opus 4.8New
claude-opus-4-8
top-tier reasoning, coding, agentic
200K
$4.00$5.00
$20.00$25.00
−20%
Claude Opus 4.6New
claude-opus-4-6
deep reasoning and coding
200K
$4.00$5.00
$20.00$25.00
−20%
Claude Sonnet 4.6New
claude-sonnet-4-6
balanced mid-tier, fast & capable
200K
$2.40$3.00
$12.00$15.00
−20%
Claude Haiku 4.5New
claude-haiku-4-5
fast, low-cost, high-volume tasks
200K
$0.80$1.00
$4.00$5.00
−20%
Whisper Large V3 TurboNew
whisper-large-v3-turbo
fast audio transcription via /v1/audio/transcriptions
Audio
$0.0004/min$0.000667/min
—
−40%
FLUX.2 ProNew
flux.2-pro
flagship image generation
—
—
$0.027/img$0.031/img
−13%
FLUX.1 SchnellNew
flux.1-schnell
fast, high-volume image generation
—
—
$0.0025/img$0.003/img
−17%
Gemini 2.5 FlashNew
gemini-2.5-flash
multimodal chat, 1M context
1M
$0.255$0.30
$2.125$2.50
−15%
Gemini 2.5 Flash ImageNew
gemini-2.5-flash-image
image generation
1M
$0.255$0.30
$25.50$30.00
−15%
Gemini 2.5 Flash LiteNew
gemini-2.5-flash-lite
high-volume cheap tasks
1M
$0.085$0.10
$0.34$0.40
−15%
Gemini 3 Flash PreviewNew
gemini-3-flash-preview
next-gen flash reasoning
1M
$0.425$0.50
$2.55$3.00
−15%
Gemini 3 Pro Image PreviewNew
gemini-3-pro-image-preview
pro-grade image generation
1M
$1.70$2.00
$102.00$120.00
−15%
Gemini 3.1 Pro PreviewNew
gemini-3.1-pro-preview
flagship reasoning, 1M context
1M
$1.70$2.00
$10.20$12.00
−15%
Gemini 3.5 FlashNew
gemini-3.5-flash
next-gen Flash GA, 1M context
1M
$1.275$1.50
$7.65$9.00
−15%
Gemini 2.5 ProNew
gemini-2.5-pro
pro mid-tier reasoning, 1M context
1M
$1.0625$1.25
$8.50$10.00
−15%
Gemini 3.1 Flash LiteNew
gemini-3.1-flash-lite
newest low-cost workhorse, high volume
1M
$0.2125$0.25
$1.275$1.50
−15%

Text-model comparisons use OpenRouter, Together AI, and Fireworks AI. Image and audio rows show QSP list pricing as of May 2026.

Side-by-side pricing vs every competitor

vs OpenRouter
20% cheaper
vs Together AI
71% on R1
vs Fireworks
75% on R1
vs DeepInfra
Lower list
vs OpenAI
Up to 30x
Coding
DeepSeek V3 for tool-calling agents →
Reasoning
DeepSeek R1 for math & algorithms →
Long context
Qwen3.5-35B-A3B for 262K RAG →
See all comparisons →
Calculator

How much would you save?

Plug in your monthly usage — see the cost on QSP vs every competitor.

Common totals (10:1 input/output):
Thinking model — output token counts include the reasoning trace, which is typically 3-10× the visible reply.
1M
300K
QuickSilver Pro
$0.13cheapest
OpenRouter
$0.16+25%
OpenAIclosed model analog
$0.33+2.6×
QSP saves 3¢/month vs OpenRouter (20% cheaper).
CLIqsp

Built for terminals and AI agents. --json output with stable exit codes — Claude Code, Cursor, Aider can call it without parsing HTML.

PyPIGitHubQuickstart →
FAQ

Common questions

An OpenAI-compatible HTTP API for 9 top open-source LLMs — DeepSeek V4 Flash & Pro, V3, R1, Qwen 3.7 Max, 3.6 Plus, 3.6 + 3.5-35B-A3B, and Kimi K2.6. Point the official OpenAI SDK at our base URL and get the same chat-completions interface, 20% below competing resellers.

V4 Flash is DeepSeek's newest model (released April 2026): ~74% cheaper output than V3, 1M context vs 128K, and thinks by default (chain-of-thought reasoning) — so a one-token "Hi" can return ~175 reasoning tokens. For V3-style cheap chat without the thinking overhead, pass `reasoning: { enabled: false }` in the request body. Existing V3 keeps working unchanged.

20% below the public per-token rates at OpenRouter, Together AI, Fireworks AI, and DeepInfra on the same open-source models. V4 Flash: $0.08 / $0.16. V4 Pro: $0.348 / $0.696. V3: $0.16 / $0.616. R1: $0.56 / $2.00. Qwen 3.7 Max: $2.00 / $6.00. Qwen 3.6 Plus: $0.26 / $1.56. Qwen 3.6: $0.12 / $0.80. Qwen 3.5: $0.111 / $0.80. Kimi K2.6: $0.584 / $2.79. We don't serve closed models (GPT-4, Claude).

Yes. Change base_url to https://api.quicksilverpro.io/v1 in the official openai Python / Node / Swift SDKs. Streaming, tool calling, json_schema strict mode, and usage.cost accounting all work out of the box.

Launch bonus: any first purchase between $5 and $50 doubles. Pay $5 and get $10. Pay $50 and get $100. One-time bonus on your first credit purchase. After that it's standard pay-as-you-go.

See all questions

Start saving on inference today

Create an account, buy credits, get your API key in 30 seconds.

Get API Key