Separate bind and client addresses, include Ollama's response body in HTTP failures, and strip inline <think> blocks from complete and streamed replies.
Ported from downstream development. Four independent defects.
1. The memory dump was unrestorable. iterdump() serializes sqlite_vec virtual
tables as a raw INSERT INTO sqlite_master(...) followed by inserts into a
table the replaying connection cannot see, so replaying memory.db.sql died
on "no such table: vec_messages" and left ZERO tables behind. dump_db() now
loads the vec0 extension and filters the derived vec tables out of the
iterdump stream, matched on each statement's target table rather than as a
substring - a chat message whose text mentions vec_messages is an
INSERT INTO "messages" and has to survive.
compare() reported an unreadable dump as "diverged", which read like a real
verdict and made both guards refuse backup AND restore, locking the machine
out of syncing in either direction. Unreadable is now its own verdict.
_extra() compared updated_at against a "" default, but the column is REAL,
so the comparison raises TypeError on the first conversation the other side
lacks - exactly the case it counts. It tests membership first now. The
direction test declared updated_at TEXT, which is why this survived: the
test compared str to str while the field compared str to float.
2. The memory curator invented facts. It attributed the ASSISTANT's words to
the user, wrote absence claims read off the existing-memory block, and added
judgements ("favorite") the user never used. The prompt now scopes the USER
line as the only source, and two deterministic guards drop absence claims
and facts whose distinctive tokens appear nowhere in the user's message -
prompt wording alone did not hold on a 7B curator.
3. _best_vulkan_device scored Mesa's llvmpipe above an integrated GPU, pinning
Ollama to a software rasterizer advertising 31 GiB of "VRAM" - CPU inference
with Vulkan overhead on top. Software rasterizers are dropped.
4. Models.jsx compared catalog names to installed names literally, but Ollama
resolves a bare name to ":latest", so an untagged entry (nomic-embed-text)
read as missing forever and the Required gate never opened. Chatbot.jsx
fetched the model list once on mount although App keeps the page mounted
behind display:none, so a newly pulled model never appeared in the picker
until a full browser reload.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Settings "Auto model routing" picks which installed model fires for chat vs
coding intent when no model is pinned (auto_chat_model / auto_code_model).
_auto_select_model honors the remap; _MODEL_PREFERENCE["code"] prefers real
coder models first.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
- install-windows.ps1: warn about a leftover profile-based ncp() that
would shadow ncp.cmd; the "press any key to close" footer now skips
the wait when stdin is redirected (was hanging indefinitely) and
exits cleanly instead of Stop-Process when it owns the window; pulls
nomic-embed-text alongside the chat/memory models.
- management/ncp.py: ncp start / start -b bring Ollama up automatically;
longer timeout + real error message on a slow model warm.
- synapse/nexus_config.py: DEFAULT_EMBED_MODEL, single source of truth
alongside DEFAULT_CHAT_MODEL/DEFAULT_MEMORY_MODEL.
- synapse/ollama_manager.py: is_available() cached instead of spawning a
process per /status poll; is_running() timeout dropped 2s -> 0.5s so a
healthy backend stops reading as dead; embed() reads the new default
instead of a hardcoded string; de-duplicated serve-env setup.
Ported via bin/publish.sh from NexusOS-jon.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Athena
enderofwingsandClaude Sonnet 5
GitNexus · part of the Nexus stack