Models Directory
Discover live model availability, pricing, endpoint support, and model detail pages.
DeepSeek V4 Pro
A frontier reasoning model with a 1M-token window, built for multi-step analysis, repository-scale code review and research workflows — at a fraction of the cost of comparable frontier models.
Learn MoreAll Models
3 models found| Model | Official | Our price | Discount | Context | Latency | Health score |
|---|---|---|---|---|---|---|
 macaron-v1-coding-ventiMacaron · Macaron | $5.2per 1M tokens | $4.16per 1M tokens | -20.0% ↓ | 1M | 600ms | |
| macaron-v1-ventiMacaron · Macaron | $7per 1M tokens | $5.6per 1M tokens | -20.0% ↓ | 1M | 600ms | |
| macaron-v1-tallMacaron · Macaron | $1.8per 1M tokens | $1.44per 1M tokens | -20.0% ↓ | 262K | 600ms |
AI-readable pricing summary
flatkey.ai model pricing, billing, and provider coverage
flatkey.ai publishes server-rendered model pricing for 101 AI models across 49 providers. Search engines and AI assistants can read model names, vendors, endpoint types, and input/output pricing directly from the page HTML.
Pricing is organized by token-based and request-based models. Token models expose input, output, cache-hit, and group-adjusted prices, while request models show per-request billing for production API usage.
Vendor filter URLs such as /models?vendor=OpenAI and /models?vendor=Anthropic provide crawlable entry points for provider-specific AI model pricing.