# Roadmap and Boundaries

> The roadmap focuses on proof-first expansion: stabilize the .slm runtime, improve tokenizer coverage, stage candidate models, deepen adapters, and expand browser-local composition only after validation.

- Site: TinyRustLM
- Canonical: https://TinyRustLM.mirust.com/roadmap/
- Version: 1.3.0
- Updated UTC: 2026-06-29T22:59:10Z

## Near-term

Keep the site honest about current implementation: one local runtime path, strict .slm validation, q8/q4 models, deterministic fixtures, adapter stack proof, selector registry routing, and clear diagnostics.

- Improve public docs
- Keep llms files aligned
- Publish support matrix
- Add evidence-oriented examples

## Middle-term

Expand model candidates and tokenizer handling where the runtime can prove compatibility. Add more concrete demonstrations of Pythia, Llama-style, low-vocabulary, tied-output, and BPE paths only when conversion and smoke evidence exist.

- Tokenizer matrix
- Candidate model table
- Browser demo evidence
- Evaluation sidecars

## Future lanes

Future work can explore multi-model routing, richer adapter ecology, OPFS persistence, Web Workers, WebGPU comparison, fixed-point deterministic math, and skill-package routing. Each lane needs a manifest, validator, test, and public-claim boundary.

- Multi-model routing
- OPFS cache
- Worker isolation
- WebGPU comparison
- Fixed-point fallback