Skip to content
NewMS Lattice found both sealed targets blind, in 32 minutes, for under 20 cents.See the proof →
Advancing Bits to Master Atoms

Search in bits.
Build in atoms.

The materials discovery engine your AI agent drives. Name what the material must do. MS Lattice finds the crystal, computes what it can do, verifies it with first-principles physics and proves it blind.

MS Lattice is the engine. Your AI agent is the driver. It runs on MatterSpace’s proprietary runtime intelligence, which gets sharper with every evaluation.

Across an orderly grid of matte graphite candidate crystals on a slate platform, one crystal of satin silver and cobalt-blue spheres rises at the center into a precise cobalt-blue unit-cell frame.Across an orderly grid of matte graphite candidate crystals on a slate platform, one crystal of satin silver and cobalt-blue spheres rises at the center into a precise cobalt-blue unit-cell frame.

One sentence in. A shortlist out.

Five steps from your words to a ranked shortlist, with the proof attached. A miss stops on the spot.

Step 1 of 5: Goal

Say what the material must do.

Hundreds of the right candidates, not millions of maybes.

Generating millions of structures is easy. Making them is not. MS Lattice holds your requirements inside the search and pushes your preferences as far as they go, so compute goes to contenders, never to a haystack.

Your limits act before the compute does.

  • Requirements, inside the search. Proposals that break your limits are turned away before a structure is built.
  • Preferences, pushed as far as they go. Nice-to-haves are pushed on top of your must-haves, never at their expense.
  • A shortlist your lab can make. Ranked candidates, each with its record, instead of a haystack.
  • Compute spent only on contenders. About a third of a cent per candidate.
Step 1 of 4: Brief

MS Lattice starts from your requirements, not a blind pile of structures.

Proven blind.

We sealed the targets away and handed over nothing but the brief. Both came back, down to the arrangement of their atoms. Four compounds it had never seen, CaO, SrF₂, BaTiO₃ and Ni₃Al, came back too, to within a few hundredths of an ångström.

sealed targets found blind
2 of 2
unseen compounds rebuilt, within 0.016–0.037 Å
4 of 4
right structures ranked first, from a formula alone
13 of 15
candidates rebuilt from their record
73 of 73

Four modes. One engine.

One engine, four kinds of question: from a target with nothing to start from to a benchmark whose answers stay sealed.

What the engine sees
Your target, nothing more

Open Discovery

You give it
A domain, a goal preset or your own targets, and a budget.
It does
It searches the whole composition space.
You get
A ranked, diverse set of candidates, each with its record.

Best for: New materials, from the target alone.

Validation you can inspect.

A score is a claim. MS Lattice gives you the proof behind it: every candidate is judged in stages, and every check it passes stays on its record.

Every verdict comes with its reasons.

Step 1 of 5: Valid

The structure is valid.

Eleven material domains.

Batteries, magnets, coatings, catalysts, ceramics. Each domain speaks its own language of targets and limits. MS Lattice speaks all eleven.

  • Battery cathodes

    More energy, less cobalt.

    Layered NMC oxides · Layered NCA oxides · Spinel oxides · Polyanion frameworks · Sodium-ion layered oxides · Prussian blue analogues · Magnesium-ion spinels

  • Solid electrolytes

    Fast ions, in a stable solid.

    Garnets · NASICON-type conductors · Sulfides · Halides · Oxide-ion conductors · Proton conductors · Ceramic–polymer composites

  • Catalysis

    From single active sites to whole surfaces.

    Single-atom alloys · Alloy surfaces · Oxide-supported catalysts · Acid OER anodes · ORR alloys · CO₂ reduction surfaces

  • Magnets

    More field, fewer rare earths.

    High-performance NdFeB · Reduced-rare-earth NdFeB · Rare-earth-free MnAl · Hexaferrites · Iron–silicon · Iron–cobalt · Spinel ferrites

  • Photovoltaics

    Absorbers for the next solar cells.

    Halide perovskites · Double perovskites · Chalcohalides · Chalcogenide absorbers

  • Thermoelectrics

    Waste heat, put to work.

    Chalcogenides · Skutterudites · Half-Heuslers · Oxide thermoelectrics

  • High-entropy alloys

    Strength from many elements.

    Transition-metal alloys · Corrosion-resistant alloys · Lightweight alloys · Refractory alloys

  • Superconductors

    Five families, one search.

    Cuprates · Iron-based pnictides · Bilayer nickelates · MgB₂-type compounds · High-pressure hydrides

  • Thermal barrier coatings

    Hotter engines, protected parts.

    Stabilized zirconia · Rare-earth zirconates · High-entropy oxides · Environmental barrier coatings

  • Hard coatings

    Harder, hotter, longer-lasting.

    Nitrides and carbides · Diborides · MAX phases

  • Ultra-high-temperature ceramics

    For the hottest places in flight.

    Rock-salt carbides · Diboride composites · High-entropy carbides · High-entropy diborides

For ceramic–polymer composites, the solid-electrolyte domain searches for the ceramic filler.

Your AI agent just became a materials scientist.

MS Lattice is the engine. Your AI agent is the driver. Your agent states the need, plans the search, runs it for minutes or for weeks, and reads back the proof.

Describe. Configure. Launch. Read.

  • One connection, 120 tools
  • Campaigns for minutes or for weeks
  • Any agent that speaks MCP
Step 1 of 4: Describe

Your agent states the need. MS Lattice names the domain that serves it.

Two ways to work.

License MS Lattice for your own team and agents, or let Norck take your target all the way to a finished part.

  • License MS Lattice

    Your team. Your agents. Our engine.

    Every domain, every mode, from your own AI agent over MCP, the REST API, the Python SDK or the command line.

    License MS Lattice →
  • Discovery with Norck

    From a target to a finished part.

    Norck runs MS Lattice, then synthesizes, validates and manufactures the winner, all the way to your finished part.

    Discovery with Norck →

From the search to your lab.

MS Lattice does the search and the proof. Your lab starts from the shortlist.

  • Ready for the lab

    A ranked shortlist, every candidate with its record.

  • Exported as structures

    A standard structure file for every candidate.

  • Proven blind

    Both sealed targets found, and four compounds it had never seen rebuilt, without being shown.

  • On to manufacturing

    Through Norck, a discovery becomes a finished part.

The first of our discovery engines.

MS Lattice is the first of MatterSpace’s discovery engines. More are on the way, for every problem in science and engineering whose space of possibilities is too large to search one try at a time.

Questions about MS Lattice.

Straight answers.

What is MS Lattice?

The materials discovery engine your AI agent drives. Name what the material must do. It finds the crystal, computes what it can do, verifies it with first-principles physics and proves it blind.

Which materials does it cover?

Eleven domains, from battery cathodes to ultra-high-temperature ceramics.

What are the four modes?

Open Discovery finds a material from a target alone. Prototype-Guided Optimization makes a good material better. Guided Rediscovery shows the engine reaching a material you know. The Blind Rediscovery Benchmark seals the answer and checks whether it comes back.

Can my own AI agent use it?

Yes. One connection, any model that speaks MCP.

Can MatterSpace take a discovery to production?

Yes. Through Norck, from a target to a finished part.

Stop guessing which crystal to make. Ask for it.

Tell us the domain and what the material must do. License MS Lattice for your team, or let Norck take the discovery all the way to finished parts.