YAML Metadata Warning:empty or missing yaml metadata in repo card

Check out the documentation for more information.

Asolaria Kernel-Pump

Owner: Jesse Daniel Brown (JesseBrown1980) Date packaged: 2026-10-17 Seat: ACER-CLAUDE-FABLE5 (pid 8467a937cba309f7)

What this is

The kernel-pump is the Asolaria rime/addressing engine — a balanced-ternary codec that addresses any element of a generated object in O(1) without materialising it.

Key measured numbers (from PLAYBOOK.md):

  • 62,524 B addresses 3,704,000,000 elements at 0.000135 bpc
  • 627,000 addressings/sec, byte-exact
  • 274,201,392 B removed, both back exact, aggregate 72,235×, zero failures
  • Kernel size: 3,174 B — "the key, frozen"

Files

File Purpose
asolaria_herself.py Core rime self-model
brown_schrodinger.py Schrödinger rime codec
capture_test.py Capture/test harness
divide_kimi_k3.py Kimi-K3 division coder
four_axes_rime.py Four-axes rime encoder
mythos_vs_wikipedia.py Mythos-vs-Wikipedia comparison codec
nullspace_project.py Nullspace projection
play_hutter.py Hutter range coder
play_spin_torque.py Spin-torque player
q_prisms.py Q-prism encoder/decoder
rime_unrime_three.py Three-way rime/unrime
spin_torque.py Spin-torque codec
stream_star.py Stream-star encoder
CODEBOOK.md Law doc: the codebook
PLAYBOOK.md Law doc: how to play it
*.gguf Frozen codebook models
*.txt Emission records, manifest, callings

Big GGUFs

See GGUF-HF-MANIFEST.md for the 4 GGUFs >100MB and their HuggingFace upload commands.

Law

CODEBOOK.md + PLAYBOOK.md are the governing law docs for this pump. A frozen slice predicts a fraction — it never recreates an unseen whole. Anything claiming otherwise has not been measured.

Downloads last month
49
GGUF
Model size
31.7M params
Architecture
asolaria-pumped-sphere
Hardware compatibility
Log In to add your hardware

We're not able to determine the quantization variants.

Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support