The Infe API Ecosystem

Precision-engineered model offerings split across two distinct layers: Infe Edge AI for instantaneous global responses, and Infe Core for heavy-duty infrastructural reasoning.

Priority Edge Network

Infe Edge AI

Deployed across 300+ global PoPs. Our edge models are tuned for sub-50ms TTFT (Time To First Token). Zero-cold-start architecture leveraging our proprietary global CDN.

300+ Global Locations

Sub-50ms Latency

Serverless Architecture

High-Margin Economy

Infrastructural Core

Infe Core

The heavy-duty tier. Optimized for raw throughput and massive reasoning. Powered by specialized inference technology clusters for the biological limit of speed.

Specialized LPU Clusters

High Reasoning Density

Ultra-High Throughput

Enterprise Grade SLA

Edge Catalogue

Pricing per 1k Tokens

Infe Pulse 8B

Language

Optimized for global edge deployment via CDN nodes.

< 50ms

50 IU

Infe Echo Edge

Speech-to-Text

High-fidelity transcription at the browser edge.

< 100ms

80 IU

Infe Vision Nano

Vision

Real-time object detection and scene analysis.

< 80ms

60 IU

Core Infrastructure

Pricing per 1k Tokens

Infe Titan 70B

Reasoning

High-parameter model running on specialized LPUs.

Sub-100ms

250 IU

Infe Titan 120B

Deep Inference

Massive reasoning capabilities for enterprise logic.

Fast-Batch

450 IU

Infe Forge Pro

Coding

SOTA code generation with 128k context window.

Sub-100ms

200 IU

Infe Vision Ultra

Multimodal

High-resolution visual reasoning and OCR.

< 150ms

300 IU

The Unit Economy

Infe.io operates on a fixed-value currency called Infe Units (IU). 1,000,000 IU is precisely calibrated to $20.00 USD. This allows for high-granularity billing across multimodal requests without complex decimal logic.

Exchange Rate

50k IU = $1.00