Pre-alpha · nothing released yet
A foundation lab.
The long-term goal is training models from the ground up. The starting point is deliberately narrower: build the evaluation we trust before we train anything we care about.
The plan
Foothills are the ground you cross before the summit. The name is the plan.
-
01
Benchmarks
Build evaluation harnesses we trust before we train anything we care about. A lab that cannot measure cannot improve.
-
02
Fine-tuning small models
Learn the full training loop — data, tokenizer, training, eval, release — at a scale where iteration is cheap and mistakes are recoverable.
-
03
Build up from there
Each rung earns the next. Scale is a consequence of competence, not a substitute for it.
Models
Elevation tracks capability tier. Continent tracks regional and language specialization.
| Glyph | Model | Continent | Elevation | Role | Released |
|---|---|---|---|---|---|
| Everest | Asia | 8,849 m | Flagship — the largest model we train | — | |
| Aconcagua | South America | 6,961 m | Spanish-language model | — | |
| Denali | North America | 6,190 m | Reserved | — | |
| Kilimanjaro | Africa | 5,895 m | Fast tier — the one anyone can walk up | — | |
| Elbrus | Europe | 5,642 m | Reserved | — | |
| Vinson | Antarctica | 4,892 m | Reserved | — | |
| Kosciuszko | Oceania | 2,228 m | Reserved — the smallest of the seven | — |
Each glyph is contoured from the mountain's elevation at a constant 1,500 m interval, so ring count is the tier by construction. Siwalik — the literal foothills of the Himalaya — is the codename series for experimental and pre-release checkpoints, and the house mark.
Code
Each released artifact gets its own repository. Apache-2.0, and nothing is published to a registry yet.
regexbench
Evaluate generated regular expressions: semantic equivalence, correctness, and ReDoS safety.
labloop
Agent-driven experiment loop: propose a change, run it time-boxed, keep it only if the metric improves.
regexleaderboard
Open code behind a regular expression leaderboard — scoring, submissions, and published results.
Roadmap
Where the work actually is, rather than where it is going.
| Phase | Focus | State |
|---|---|---|
| 0 | Repo, branding, project foundations | In progress |
| 1 | Benchmark harness and first eval suite | Next |
| 2 | Fine-tuning pipeline for small open models | Planned |
| 3 | First published model and eval results | Planned |
| 4 | Pretraining experiments | Horizon |
Principles
Measure before you train
Every model claim ships with the eval that backs it, and the harness that produced it.
Small and honest beats large and vague
A 1B model with a reproducible number is worth more than a big claim with none.
Reproducibility is a feature
Seeds, configs, and data provenance are part of the result, not an appendix to it.
Publish the negative results too
The failures are most of the signal.
If something here is useful to you, or wrong, open an issue.