For teams building with more than one model

Shared memory and routing for multi-model AI

82d projects embeddings from different providers into one stable 82D space, so you can search, route work, and migrate models without rebuilding your stack every time a vendor changes.

Start Free See How It Works
Cross-model memory Semantic routing No re-embedding migration path
45M+ verified vectors/sec
18.7× smaller vectors
41.5M public passages ready
New user path

I need the 60-second explanation

Understand what 82d replaces, what stays stable, and why it matters before the API details.

 Hands on

I want to try the API first

Start with the quick projection flow, then decide whether you need routing, memory, or public data.

 Fit check

I want to know where this fits

See the most common starting use cases: migration, shared search, and prebuilt public retrieval.
~ ~ ~

How it works

Start with one narrow job: make vectors from different models usable in the same system.

1

Project embeddings into 82D

OpenAI, Cohere, MiniLM, mxbai, and other embeddings land in one compact shared coordinate system.

2

Search and route in one shared space

Memory, retrieval, and routing stop depending on incompatible dimensions and provider-specific wrappers.

3

Keep continuity when vendors change

Your 82D coordinates stay usable when you swap models, merge teams, or add prebuilt public data.

~ ~ ~

Why teams end up needing this layer

These costs compound when search, memory, and routing depend on model-specific embeddings.

Storage bloat

1536D vectors = 6 KB each. At 10M documents, that's 61 GB of coordinates — most of it redundant dimensions you never query directly.

Routing dependency

Every query and handoff depends on model-specific APIs, wrappers, and rate limits. Your continuity and control surface live on someone else's stack.

ETL bottleneck

Want Wikipedia in your RAG? 41.5M passages. $5,000+ to embed. 63 GB to store. Months to build the ingest pipeline.

Continuity break

text-embedding-ada-002 → sunset. Re-embed everything. Pay again. Rebuild compatibility. Repeat forever.

one layer, three immediate jobs
~ ~ ~

What you get immediately

A stable memory and routing layer you can use before you adopt the broader service stack.

Continuity across model changes.

  • Cross-model memory: embeddings from OpenAI, Cohere, mxbai, nomic, MiniLM, and more land in one shared 82D space
  • Semantic routing: one representation layer for search, handoff, and mixed-model retrieval
  • Interoperability: merge datasets embedded by different teams or providers into one usable system
  • Continuity: when a provider sunsets a model, your 82D coordinates stay stable
  • 18.7× smaller: 6,144 bytes → 328 bytes per vector, reducing RAM, storage, and bandwidth costs
Try It Now
~ ~ ~
● Live Now

The Firehose

Public knowledge already projected into the same shared memory layer. Add continuity and retrieval coverage without building your own ingestion machinery.

Wikipedia — 41.5 million passages. Every article, every paragraph, projected to 82 dimensions. Semantic search across all of human knowledge in 25ms. Plug it into your RAG pipeline in minutes, not months.

Coming next: ArXiv papers, PubMed abstracts, Common Crawl domains. If it's public knowledge, we're projecting it.

41.5M passages
25ms search latency
328B per vector
$0 embedding cost
Explore Firehose →
~ ~ ~

Quick Start

Project embeddings from any model into the same memory and control layer in one API call.

Python 
from eightytwo import Client

client = Client(api_key="your-key-here")

# Works with ANY embedding model
# OpenAI 1536D, Cohere 1024D, nomic 768D, etc.
vectors_1536d = openai_client.embeddings.create(...).data
vectors_82d = client.project(vectors_1536d)
# → model auto-detected from dimension

# Or specify the model explicitly
vectors_1024d = mxbai_client.embed(texts)
vectors_82d = client.project(vectors_1024d, model="mxbai-embed-large")

# Both land in the SAME 82D consensus space
# → directly comparable, permanently yours
print(f"Size: {1536*4}B → {82*4}B per vector = 18.7x smaller")

Sign up to get your API key and endpoint URL.

cURL 
# Project vectors to 82D consensus space
curl -X POST https://api.82d.ai/project \
  -H "Authorization: Bearer YOUR_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "vectors": [[0.01, -0.02, ...1536 dims...]],
    "model": "openai-3-small"
  }'

# Response:
{
  "vectors": [[0.04, 0.10, ...82 floats]],
  "count": 1,
  "input_dim": 1536,
  "output_dim": 82,
  "processing_time_ms": 0.3
}

# List supported models
curl https://api.82d.ai/models

Paste 1536-dimensional vectors (from OpenAI, Cohere, etc.) to project to 82D.

Sign in above to enable live demo

~ ~ ~

Simple Pricing

Free during early access. Usage is monitored per account.

You own the coordinates. Nothing stored on our side unless you ask.

Output size 0.31 GB
Your cost $0.08
Re-embed with OpenAI $13,000
You save 162,500×

Early Access — Free Tier

Every account gets free monitored access to the full projection API, pre-built Wikipedia search, and a trained W matrix. Paid credit packs will be available when we leave early access.

Starter
$10
5 GB
Popular
Builder
$50
25 GB
Pro
$100
50 GB
Scale
$500
250 GB

Every account includes a trained W matrix + Wikipedia search access. Paid tiers coming soon.

~ ~ ~

Pick a starting use case

You do not need the whole platform at once. Most teams start with one of these.

RAG at Scale

Drop 41.5M pre-projected Wikipedia passages into your pipeline. Add your own embeddings from any model. One unified 82D index.

Model Migration

Moving from OpenAI to Cohere? Project both to 82D. Zero re-embedding. Zero downtime. Your existing vectors just work.

Multi-Team Search

Engineering uses mxbai. Research uses nomic. Product uses OpenAI. 82D makes them all searchable in one index.

Cost Control

18.7× smaller vectors = 18.7× less RAM, storage, and bandwidth. At scale, that's the difference between renting GPUs and not.

~ ~ ~

Beyond Projection

Projection is the foundation. On top of it: platform migration, pre-built datasets, agent interoperability, multimodal search, and managed infrastructure — nine services, one coordinate system.

Explore Services →
~ ~ ~

Your vectors. Your coordinates. Your call.

First 10 MB free. No credit card. See the math for yourself.

Start Free