Price Comparison API

GET /v1/prices

Public endpoint — no authentication required. Returns a price comparison matrix across all providers for every model in the marketplace.

Query Parameters

Parameter	Type	Description
`model`	string	Filter to a single model (e.g. `claude-opus-4.6`)

Response Format

{
  "models": [
    {
      "model": "claude-opus-4.6",
      "displayName": "Claude Opus 4.6",
      "provider": "Anthropic",
      "contextLength": 1000000,
      "providers": {
        "venice": { "input": 6.00, "output": 30.00, "cacheRead": 0.60 },
        "openrouter": { "input": 5.00, "output": 25.00, "cacheRead": 0.50, "cacheWrite": 6.25 },
        "anthropic": { "input": 5.00, "output": 25.00, "cacheRead": 0.50, "cacheWrite": 6.25 },
        "bankr": { "input": 5.00, "output": 25.00, "cacheRead": 0.50 }
      },
      "cheapest": { "provider": "openrouter", "input": 5.00, "output": 25.00 }
    }
  ],
  "providers": ["venice", "openrouter", "anthropic", "deepseek", "bankr"],
  "updated_at": "2026-04-14T11:00:00.000Z"
}

Fields

Model Entry

Field	Type	Description
`model`	string	Standard marketplace model name
`displayName`	string	Human-readable name
`provider`	string	Model creator (Anthropic, OpenAI, Google, etc.)
`contextLength`	number	Maximum context window in tokens
`providers`	object	Price data keyed by provider
`cheapest`	object	Cheapest provider for this model

Price Entry (per provider)

All prices in USD per 1M tokens.

Field	Type	Description
`input`	number	Input/prompt token price
`output`	number	Output/completion token price
`cacheRead`	number	Cache hit price (optional)
`cacheWrite`	number	Cache creation price (optional)
`reasoning`	number	Thinking/reasoning token price (optional)

Cheapest

Field	Type	Description
`provider`	string	Provider key with the lowest weighted cost
`input`	number	Input price at that provider
`output`	number	Output price at that provider

> Cheapest is determined by weighted cost: input + 3 × output. This approximates a typical chat workload where output tokens outnumber input tokens.

Examples

All models

curl https://api.surplusintelligence.ai/v1/prices

Single model

curl "https://api.surplusintelligence.ai/v1/prices?model=claude-opus-4.6"

Caching

Response is cached for 1 hour (ISR). The updated_at field indicates when the snapshot was generated.

Providers Tracked

The live price response currently includes 10 provider pricing sources. The seller allowlist has 11 provider domains; OpenAI direct is allowlisted but may not appear in the live price matrix if no current pricing rows are available.

Key	Provider
`venice`	Venice AI
`openrouter`	OpenRouter
`anthropic`	Anthropic (direct)
`openai`	OpenAI (direct)
`google`	Google (Gemini)
`deepseek`	DeepSeek (direct)
`mistral`	Mistral AI
`groq`	Groq
`together`	Together AI
`fireworks`	Fireworks AI
`bankr`	Bankr

Interactive Price Table

See the live price comparison API for the current machine-readable matrix. Treat that response as the source of truth for current provider rows and counts.