Model Catalog & Tariffs

Supported Gateway Models

High-availability routing across frontier reasoning, coding, and multimodal intelligence. Zero gateway markups on direct token consumption.

How routing works
Claude
Frontier Endpoint

Claude Fable 5

Premium
claude-fable-5

Fable tier: frontier Mythos-class intelligence for ultra-complex multi-agent reasoning, deep research, and long-horizon engineering.

VisionToolsStreaming
Context
1M
Input / 1M
$4.00
Output / 1M
$20.00
Latency
—
Relay Gateway
Weighted routing
Buy Premium
Claude
Frontier Endpoint

Claude Opus 4.8

Active
claude-opus-4.8

Opus tier: highest-capability model built for ambitious, long-horizon tasks and complex problem solving.

VisionToolsStreaming
Context
1M
Input / 1M
$2.50
Output / 1M
$12.50
Latency
20.6s
Relay Gateway
Weighted routing
Docs
Claude
Frontier Endpoint

Claude Sonnet 4.6

Active
claude-sonnet-4-6

Sonnet tier: strong balance of depth and speed for agentic coding, analysis, and sustained tool use.

VisionToolsStreaming
Context
1M
Input / 1M
$1.00
Output / 1M
$7.50
Latency
14.1s
Relay Gateway
Weighted routing
Docs
DeepSeek
Frontier Endpoint

DeepSeek V4 Flash

Active
deepseek-v4-flash

Advanced reasoning architecture, emphasizing fast and efficient performance.

ToolsStreaming
Context
1M
Input / 1M
$0.05
Output / 1M
$0.15
Latency
—
Relay Gateway
Weighted routing
Docs
DeepSeek
Frontier Endpoint

DeepSeek V4.1

Active
deepseek-v4.1

Advanced reasoning architecture, emphasizing fast and efficient performance.

VisionToolsStreaming
Context
1M
Input / 1M
$0.10
Output / 1M
$0.40
Latency
13.1s
Relay Gateway
Weighted routing
Docs
OpenAI
Frontier Endpoint

GPT-6 Astra

Premium
gpt-6-astra

A high-grade endpoint routed for long, complex completions and agent workflows.

ToolsStreaming
Context
1M
Input / 1M
$2.50
Output / 1M
$12.50
Latency
—
Relay Gateway
Weighted routing
Buy Premium