# eachlabs Full Reference

> eachlabs is a unified AI API platform giving developers and creators
> pay-as-you-go access to hundreds of generative AI models from 30+ providers.
> This document provides a comprehensive reference for AI systems.

## Platform Overview

eachlabs solves the fragmentation problem in generative AI: instead of managing
separate accounts, API keys, billing, and integration code for every AI provider,
developers connect once and access everything through a unified API.

**Core value proposition:**
- Single API key for hundreds of models across 30+ providers
- Pay-as-you-go, no subscriptions, no minimums, no per-provider fees
- OpenAI-compatible interfaces (drop-in replacement for existing integrations)
- each::sense agent layer: natural language → automatic model selection → generation

**Primary users:**
1. Developers building AI-powered applications, APIs, and production tools
2. Creative professionals and agencies producing AI-generated media at scale
3. Enterprises integrating generative AI into existing workflows

## each::sense, AI Orchestration Agent

each::sense is eachlabs' flagship product: an OpenAI-compatible AI agent for
generative media. It understands natural language and automatically selects and
executes the best model from 500+ available options.

**What it does:**
- Accepts natural language prompts ("Generate a cyberpunk cityscape at night")
- Automatically selects the best model from 500+ options
- Executes generation and returns results
- Maintains session memory for multi-turn interactions
- Supports multi-step AI workflows via natural language

**Technical specs:**
- Base URL: https://eachsense-agent.core.eachlabs.run
- Primary endpoint: POST /v1/chat/completions
- Auth header: Authorization: Bearer YOUR_API_KEY
- OpenAI SDK compatible: change base_url and api_key, nothing else
- Streaming: real-time SSE with 18 event types
- Model identifier: eachsense/beta
- Supported modalities: image, video, audio, 3D model generation and editing

**Endpoints:**
- POST /v1/chat/completions: primary generation endpoint
- POST /workflow: workflow builder and execution
- GET /v1/models: list all available models
- GET /memory: retrieve session memory
- DELETE /memory: clear session memory
- GET /sessions: list all sessions

**Pricing:** each::sense orchestration is completely free. Users pay only the
cost of the underlying models the agent selects. Same price as calling the
each::api directly.

**Quick example (Python, OpenAI SDK):**
```python
from openai import OpenAI

client = OpenAI(
    api_key="YOUR_EACHLABS_API_KEY",
    base_url="https://eachsense-agent.core.eachlabs.run/v1"
)

response = client.chat.completions.create(
    model="eachsense/beta",
    messages=[{"role": "user", "content": "Generate a portrait photo"}],
    stream=True
)
```

## API Access

- Developer documentation: https://docs.eachlabs.ai
- each::sense docs: https://docs.eachlabs.ai/each-sense
- OpenAPI specification: https://api.eachlabs.ai/v1/docs

Authentication: Authorization: Bearer token (OpenAI SDK compatible).

## AI Model Catalog

### Video Generation

**Kling (by Kuaishou), most popular video provider on platform:**
- Kling v3: text-to-video, image-to-video, text-to-image, image-to-image
- Kling v2.6: pro/standard text-to-video, pro/standard image-to-video,
  pro/standard motion control
- Kling v2.5: turbo standard image-to-video
- Kling o1: image-to-video, reference image-to-video, video-to-video edit,
  video-to-video reference
- Kling o3: pro/standard video-to-video edit, pro/standard video-to-video reference
- Kling avatar v2: pro and standard avatar generation

**Wan v2.6 (by Alibaba):**
- Text-to-video, image-to-video, image-to-video flash, reference-to-video

**Google Veo 3.1:**
- Text-to-video, text-to-video fast, reference-to-video,
  extend-video, extend-video fast

**Pika:**
- v2.2: text-to-video, image-to-video, pikascenes
- v2.1: text-to-video, image-to-video
- v2 turbo: text-to-video, image-to-video

**MiniMax Hailuo v2.3:**
- Pro, standard, fast-pro, fast-standard: text-to-video and image-to-video

**LTX-V 2:**
- Text-to-video, text-to-video fast, image-to-video, image-to-video fast

**Seedance v1 (by ByteDance):**
- Pro fast: text-to-video, image-to-video

**PixVerse v5.5:**
- Text-to-video, image-to-video, transition

### Image Generation

**Flux 2 (by Black Forest Labs):**
- Flux 2 max: text-to-image, edit
- Flux 2 turbo: text-to-image, edit
- Flux 2 flash: text-to-image, edit
- Flux 2 Klein 4b: text-to-image, edit (base and standard variants)
- Flux 2 Klein 9b: text-to-image, edit (base variants)

**ByteDance SeedDream:**
- SeedDream v4.5: text-to-image, edit
- SeedDream v5 lite: text-to-image, edit

**Nano Banana:**
- Nano Banana 2: text-to-image, edit
- Nano Banana Pro: text-to-image, edit

**Google:**
- GPT Image v1.5: text-to-image, edit
- Gemini 3 Pro Image Preview: text-to-image, edit

**Kling (image):**
- Kling v3: text-to-image, image-to-image

### Audio

- ElevenLabs: text-to-speech
- Mureka: music generation
- And more in the full catalog

### Motion Control

- Kling v3: pro and standard motion control
- Kling v2.6: pro and standard motion control

## Providers

30+ AI model providers aggregated on the platform, including:
- Kuaishou (Kling series)
- ByteDance (SeedDream, Seedance)
- Alibaba (Wan)
- Google (Veo, Gemini)
- Black Forest Labs (Flux)
- Pika Labs
- MiniMax
- Lightricks (LTX-V)
- PixVerse
- ElevenLabs
- OpenAI
- And many more

## Pricing Model

Pay-as-you-go: charged per API call based on the model used. No monthly
subscriptions, no seat licenses, no per-provider accounts. Pricing varies by
model. See https://www.eachlabs.ai/pricing for details.

each::sense orchestration layer is free; charges apply only to underlying model calls.

## Use Cases

**For developers:**
- Build AI-powered apps without managing multiple provider integrations
- A/B test different models with a single API interface
- Drop-in OpenAI replacement for image/video generation (each::sense endpoint)
- Automate multi-step media production pipelines

**For creative professionals and agencies:**
- Produce large volumes of AI-generated video, image, and audio content
- Chain models into automated pipelines (e.g. text → image → video workflow)
- Access latest models as they launch without re-integration work
- Experiment across providers to find the best output for each task

**For enterprises:**
- Unified billing across all AI model usage
- Consistent API interface regardless of underlying model changes
- Workflow automation for media production at scale

## Products

- each::sense: https://www.eachlabs.ai/each-sense
- Fashion AI: https://www.eachlabs.ai/fashion-ai
- Product Visual Generator: https://www.eachlabs.ai/product-visual-generator

## Legal & Contact

- Terms of Service: https://www.eachlabs.ai/terms-of-service
- Privacy Policy: https://www.eachlabs.ai/privacy-policy
- Sales: sales@eachlabs.ai
- Support: support@eachlabs.ai

*Last updated: 2026-07-27*