PRIMER

Research

Follow the experiment.

Primer is an active research program exploring how intelligence can continue developing after training.

Our mission is superintelligence. We believe continual development may be one of the mechanisms required to get there. Primer is our experimental platform for studying that path.

We evaluate capabilities under frozen protocols, preserve negative results, and distinguish internal benchmarks from external validation.

The progression

Learning. Development. Beyond.

  1. 01
    Learns

    Acquires knowledge created after its training ended.

    Demonstrated
  2. 02
    Remembers

    Keeps what it learns, even while it learns more.

    Demonstrated
  3. 03
    Combines

    Uses what it knows to solve harder related problems.

    Demonstrated
  4. 04
    Compounds

    Each thing it learns makes the next easier.

    Demonstrated
  5. 05
    Abstracts

    Turns its own solutions into reusable tools.

    In progress
  6. 06
    Invents

    Creates new ideas from old ones.

    Ahead
  7. 07
    Transfers

    Carries a method from one field into another.

    Ahead
  8. 08
    Directs itself

    Chooses what to learn next.

    Ahead
  9. 09
    Discovers

    Produces verified knowledge nobody had before.

    Ahead
  10. 10
    Improves how it learns

    Each generation learns faster than the last.

    Mission

Method

Ambitious mission.
Conservative claims.

Post-freeze

Test on material created after training, so success cannot come from memory.

Measured

A capability is claimed only after it has been measured.

Labelled

Internal results are called internal until they are validated externally.

Falsifiable

Negative results matter.

The mission is superintelligence.

The research standard is evidence.

Current research areas

  • Post-training learning
  • Persistent knowledge
  • Knowledge transfer
  • Generalization
  • Scaling
  • Evaluation

Public research will focus on

  • Measured capabilities
  • Benchmark methodology
  • Externally validated results
  • High-level scientific findings

Core implementation techniques, training recipes, and proprietary architecture remain confidential.

Results published on this site come from internal research evaluations. External validation is ongoing.

Read the technical reportsoon