Trained adapters published, private to their owners like the knowledge bases they come from

An adapter now has an owner (org, team, project, user or key): only callers
there see it, and to anyone else it is a model that does not exist, absent
from /v1/models too. A training job's owner defaults to the one collection
its pairs come from, must be able to read every collection they come from,
and pairs from a knowledge base are never made public. ctl train publish
keeps the GGUF or PEFT files on the mirror, which accepts uploads now, and
sets the owned adapter after the usual architecture check.

The knowledge bases' cache_salt follows the collections searched rather
than the organization, so two teams of one organization no longer share
cached passages.

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
This commit is contained in:
Ilker ManapandClaude Opus 5.5 committed 2026-10-08 21:30:50 +02:00
1 parent 964b35ae65
commit 85025dfb03
29 files changed
+1049 -84

No files matched your search

+20 -1
View File
@@ -1005,7 +1005,7 @@ a request without the header, `none` turns it off, an explicit header with
k=2 gives two sources and no more, llmgw-kb lists owners by path. Left: returned passages
are not audited per request (the header and the kept request record them).
### F4. Adapter training (L, after F1–F3)
### F4. Adapter training (L, after F1–F3) — done 2026-10-08 (branch `adapters-f4`)
A run-to-completion job placed by the control plane like a replica and
supervised by the agent like an engine, in a Python environment the agent
installs (TRL + PEFT, QLoRA). Data: Q/A pairs generated from a collection by
@@ -1026,6 +1026,25 @@ facts come from F1–F3; a bigger base on a free card would do better but not
change that. The training itself fits a small slot on a shared card, which
the job design can rely on for small bases.
Built: `ctl kb dataset` (POST /api/v1/kb/collections/{id}/dataset, NDJSON
with each pair's collection); train_jobs/train_artifacts (migrations 0037,
0038 owner); internal/control/train (queue, one job per node, artifacts in
1 MiB pieces, a node that goes fails its job); internal/agent/train (the
newest vLLM's python3 + `uv pip install --no-deps --target` peft 0.21.2 and
accelerate 1.15.0; llama.cpp's conversion code for the installed release,
untarred from its source; the card with the most free memory, if enough;
runs as llmgw-engine); `ctl train start|list|show --follow|cancel|rm|
download|publish`; the mirror's PUT /v1/models/{model}/{version}/{file};
publish uploads the GGUF (llama.cpp) or PEFT files (vLLM) and sets the
adapter after checkAdapters. Owners: catalog.Adapter.Owner, enforced in the
router (404, hidden from /v1/models); a job's owner defaults to its single
collection's owner, must be able to read every collection the pairs came
from, and pairs from a knowledge base are never public. Tried on the PoC:
dataset of 90 pairs in 89 s; a job trained beside the 35B (peak 2214 MiB,
138 steps in about a minute) and sent back PEFT and GGUF.
Left: a job dies with its agent's stream (no resume); no console screen for
training yet; quality gating of a published adapter is by hand (ctl eval).
---
## D. Facts worth knowing before starting