Skip to content

Models & Pricing

Last updated: August 2026

All models are accessible via a single API key. Pricing is pay-per-use with no monthly minimums.

Data Freshness

Model availability and pricing change frequently. The tables below are representative examples. Always use GET /v1/models for real-time, authoritative data. If a model appears here but returns model_not_found, it may have been deprecated or renamed.

LLM Models

All LLM models support chat and streaming by default. The Capabilities column below only lists additional differentiating features.

ModelVendorContextInput ($/M tokens)Output ($/M tokens)Capabilities
openai/gpt-4oOpenAI128K$2.50$10.00vision, tools, json_mode
openai/gpt-4o-miniOpenAI128K$0.15$0.60vision, tools, json_mode
openai/o3OpenAI200K$2.00$8.00tools, reasoning
anthropic/claude-sonnet-4-20250514Anthropic200K$3.00$15.00vision, tools, cache
anthropic/claude-haiku-3-5Anthropic200K$0.80$4.00vision, tools, cache
google/gemini-2.5-proGoogle1M$1.25$10.00vision, tools
google/gemini-2.5-flashGoogle1M$0.15$0.60vision, tools
deepseek/deepseek-chatDeepSeek128K$0.14$0.28tools
deepseek/deepseek-reasonerDeepSeek128K$0.55$2.19tools, reasoning
meta/llama-4-maverickMeta1M$0.20$0.60tools
mistral/mistral-largeMistral128K$2.00$6.00tools
qwen/qwen3-235bAlibaba128K$0.80$2.40tools
minimax/minimax-01MiniMax1M$0.20$1.10tools

Image Models

ModelVendorBase PriceMax ResolutionSpeed
flux/schnellBlack Forest Labs$0.0031024×1024~2s
flux/proBlack Forest Labs$0.052048×2048~10s
flux/devBlack Forest Labs$0.0251024×1024~8s
ideogram/v3Ideogram$0.042048×2048~8s
recraft/v3Recraft$0.042048×2048~5s
openai/dall-e-3OpenAI$0.041792×1024~15s
google/imagen-4Google$0.042048×2048~6s

Video Models

ModelVendorBase PriceMax DurationResolution
kling/v1-5Kuaishou$0.2810s1080p
kling/v2Kuaishou$0.3510s1080p
minimax/video-01MiniMax$0.306s1080p
runway/gen4Runway$0.5010s1080p
luma/ray-2Luma$0.305s1080p
vidu/vidu-2Vidu$0.258s1080p

Audio Models

ModelVendorPricingFeatures
elevenlabs/multilingual-v2ElevenLabs$0.30 / 1K charsTTS, multilingual, voice cloning
openai/tts-1-hdOpenAI$0.030 / 1K charsTTS, 6 voices
openai/whisper-1OpenAI$0.006 / minSTT, multilingual

Embedding Models

ModelVendorPrice ($/M tokens)DimensionsMax Input
openai/text-embedding-3-largeOpenAI$0.1330728191 tokens
openai/text-embedding-3-smallOpenAI$0.0215368191 tokens
voyage/voyage-3Voyage AI$0.06102432000 tokens

Pricing Rules

Base Pricing

  • LLM: Input tokens and output tokens billed separately, per million tokens
  • Image/Video/Audio: Flat per-generation price (no token counting)
  • Embedding: Per million input tokens
  • No monthly minimums, no commitments

Caching & Discounts

Some LLM providers offer prompt caching for repeated context:

ProviderCache WriteCache Read (savings)
Anthropic1.25× input price0.10× input price (90% off)
OpenAI1× input price (free)0.50× input price (50% off)
Google1× input price0.25× input price (75% off)

Caching is automatic when supported — SandBase passes through the provider's cache behavior. Check the usage field in responses for cache_read_input_tokens and cache_creation_input_tokens.

Volume & Enterprise

Contact [email protected] for:

  • Volume discounts on high-throughput usage
  • Committed spend agreements
  • Custom rate limits

Query Pricing Programmatically

Get all models

bash
curl https://api.sandbase.ai/v1/models \
  -H "Authorization: Bearer sk-sb-YOUR_KEY"

Response:

json
{
  "data": [
    {
      "id": "openai/gpt-4o",
      "type": "llm",
      "vendor": "OpenAI",
      "context_window": 128000,
      "max_output_tokens": 16384,
      "pricing": {
        "input_per_million": 2.50,
        "output_per_million": 10.00,
        "cached_input_per_million": 1.25
      },
      "capabilities": ["chat", "vision", "tools", "json_mode", "streaming"],
      "status": "active"
    },
    {
      "id": "flux/schnell",
      "type": "image",
      "vendor": "Black Forest Labs",
      "pricing": {
        "base_price": 0.003
      },
      "capabilities": ["text_to_image"],
      "status": "active"
    }
  ]
}

Get a specific model

bash
curl https://api.sandbase.ai/v1/models/anthropic/claude-sonnet-4-20250514 \
  -H "Authorization: Bearer sk-sb-YOUR_KEY"

Check cost after generation

bash
curl https://api.sandbase.ai/v1/tasks/{task_id}/cost \
  -H "Authorization: Bearer sk-sb-YOUR_KEY"

Response:

json
{
  "task_id": "task_abc123",
  "model": "openai/gpt-4o",
  "cost_usd": 0.000325,
  "usage": {
    "prompt_tokens": 24,
    "completion_tokens": 89,
    "cache_read_input_tokens": 0
  },
  "created_at": "2026-08-02T12:00:00Z"
}