NanoChat.NET — UI + MCP surface

What the desktop UI and MCP server actually expose of that same engine.

16 capabilities 7 sections extracted from UI + MCP surface (WinApp) source: D:\Dev\Lab\NanoChat.NET

Build against it

# give your agent the catalog — no account, no install
claude mcp add provespec -- npx -y provespec-mcp

# then, from any agent session
spec_get(slug: "nanochat-net-ui")                → the capability tree
spec_checklist(slug: "nanochat-net-ui")          → the questions to answer
spec_grade(slug: "nanochat-net-ui", grades: [...])  → parity % + the gap list

# or just fetch it
curl https://provespec.com/catalog/nanochat-net-ui/spec.json

Grade every capability below as yes · partial · no · na, with a file reference as evidence. Whatever you leave ungraded counts as missing — the gap list is the work queue.

Or just paste this to your agent

Use the ProveSpec MCP server. Fetch the "nanochat-net-ui" spec with
spec_get, then walk spec_checklist and grade this repository against it —
one verdict per capability, each with a file reference as the note.
Call spec_grade with the results and give me the gap list as the plan.

If the server is not registered, add it first:
  claude mcp add provespec -- npx -y provespec-mcp

The specification

Tokenizer 2

✗ Train tokenizer training runs via CLI only; no UI action found (2026-08-01 snapshot)
✓ Inspect vocabulary & merges Tokenizer explorer: live retokenize, vocab + merge-history browsers

Data 1

✓ Register & inspect corpora Datasets activity (14 VMs) + MCP datasets_source_corpora/add

Training 4

✓ Pretraining runs Training workflow catalog: run cards, stage progress, ETA + MCP training_*
✓ Task/SFT training (5 kinds) TrainingActionCatalog actions incl. dataset chooser
✓ Resume from checkpoint *-resume actions with compatibility checks
✗ RL training loop complete at model level; no UI activity, no rl lineage (confirmed by Mind v8)

Evaluation 3

✓ BPB / CORE / GSM8K / ARC / MMLU Benchmark setup & results UI + bpb_pilot_* MCP tools
✓ Evaluation history & drilldown data Evaluation drilldown activity: per-example prompts, losses, verdict margin
✗ Inference performance benchmark engine InferenceBenchmark not surfaced as a UI activity (snapshot)

Inference 2

✓ Chat sessions (streamed, tool-use) Chat activity with model picker, streamed replies + MCP chat_*
✓ Task runtime (generate/classify/rank/embed) Task Runtime deployment activity (typed text against active checkpoint)

Checkpoints & models 3

✓ Checkpoint inventory Checkpoints observatory: lineage > tag > step from real disk
✓ Registry: deploy & rollback Models activity: Load/Inspect/Register/Activate/Rollback
✗ Upstream checkpoint import importer exists engine-side; no import UI action found (snapshot)

Internals 1

◐ GPU/tensor/kernel state GPU dashboard, tensor workspace, 12 renderers — but ~4/99 render entry points read the real model

Download spec.json All 53 specs