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
    Learning

    Can the system acquire knowledge generated after its original training?

    Demonstrated in internal evaluations
  2. 02
    Memory

    Can verified knowledge persist through later learning?

    Demonstrated in internal evaluations
  3. 03
    Composition

    Can previous learning become a useful building block for harder new learning?

    Demonstrated in internal evaluations
  4. 04
    Development

    Can earlier learning make future learning more powerful, autonomously?

    Demonstrated in internal evaluations
  5. 05
    Generalization

    Can the learning architecture extend across broad domains and concept classes?

    Active research
  6. 06
    Compounding intelligence

    Can learning and improvement compound beyond human capability across broad domains?

    Mission

Method

Ambitious mission.
Conservative claims.

Post-freeze

Test knowledge generated after checkpoints are frozen.

Verified

A model output is not automatically considered knowledge.

Persistent

Learning should survive beyond one interaction.

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