- download: resumed downloads checked against a pinned SHA-256.
- install: uv 0.11.20, the worker/brain venv (Python 3.12, the new
src/voice_worker/requirements-brain.txt), the MeloTTS venv (Python 3.11,
PyTorch cu128), Whisper large-v3-turbo, the Korean MeloTTS voice and
Strata, the M0 kit's recipe; markers per step and version, cancel, retry,
everything (caches, temp, HF and NLTK data) inside the data folder, the uv
cache cleaned and the worker settings pointed at the venvs when done.
- The services give the worker the same environment once installed.
- The wizard's model step shows and drives the install.
Installed from an empty folder on Linux and started the services on it:
worker connected, STT and TTS ready, nothing written outside the folder.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- admin_api: the one entry point of the screens (app status, data folder,
PC check, Discord token check and pickers, settings, service control and
the dashboard, logs query, error report and zip export).
- discord_setup: token, Message Content and invite link checks over REST.
- app/ui: first-run wizard, dashboard, five settings screens and the log
screen in plain HTML/CSS/JS (dark theme after DESIGN.md, Korean text).
- jarvis-admin-dev: a loopback, token-guarded development server so the
screens run in a browser against the real commands.
Checked in a browser with the real test bot and voice worker: wizard
steps, start/restart/stop, settings save, chat channels, logs and export.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- control: one Controller starts, stops and restarts the services (stopped,
starting, running, stopping, failed); a failed start leaves nothing
running and keeps its error with secrets masked. The app binary uses it.
- dashboard: built from the core's events alone: the bot's tag, the voice
channel, live captions (screen only, cleared on leaving or stopping), the
last 20 voice answers split into stages with medians (matched by the new
AnswerAsked event), and missed requests without text.
- system: memory in use and per NVIDIA GPU its use and VRAM (nvidia-smi).
- DiscordCore::voice_members, WorkerClient::hello, the conversation's missed
requests shared with the dashboard.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- datadir: the data folder layout, the location file (JARVIS_DATA_DIR wins),
free space for the wizard.
- settings_edit: the fields the five settings screens edit; apply writes only
non-defaults, keeps unknown keys and writes nothing unless the app's own
check passes. A memory section reaches the worker's hello.
- secrets: config/secrets.dat, DPAPI for the Windows user; the services take
the bot token from it before JARVIS_DISCORD_TOKEN.
- logs: the log hub behind the log screen (tabs, daily files kept 14 days,
the recent error report, a zip under 10 MB, secrets and token-shaped text
masked); the log macros and the worker's output feed it.
Spec: app/src/admin.spec.md.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Every <parameter> value was passed as text, so a delete confirmed with
confirm=true reached chatChannels as "true" and asked again. Values now take
the type the tool's input schema declares (boolean, integer, number, array,
object); a value that does not fit stays text. The roadmap's eval summary now
names the reply-length failure apart from the later passing runs.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
live_round_trip.rs runs the app's services with the real worker and brain,
plays recorded questions from a second bot and times the final, the
answer text, playback start and the first audible frame after the speech
ends; a second worker session streams nine more voices for the load run.
Roadmap: first M4 step 4 numbers (median 3.5 s alone, 4.4 s under load).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A pause after the name ("자비스, ... 지금 몇 시야?") is longer than the
worker's 300 ms end silence, so the wake word ended its own utterance:
Jarvis answered the bare name and ignored the question that followed (no
wake word). Found on real Discord. A voice utterance that is only a wake
word is now joined to the same speaker's next final when that utterance
is running or starts within call.followUpMs (default 1500); otherwise the
wake word alone is the request after that time. Stops drop it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The voice and chat lanes let a person's voice reply start while their chat
reply was still running. The chat delete consent counts "the very next
message" by turn number, so a message that started before the delete was
asked could pass as the confirmation, and the person's dialogue memory
was updated out of order.
- A reply takes its asker's lock, then its lane's: one person's replies
stay one at a time in the order asked, different people's voice and
chat answers still overlap, and a reply waiting for its asker holds no
lane. Test: a person's voice question starts only after their chat
answer.
- Specs and roadmap: the lanes are per person; the roadmap no longer says
voice answers wait behind chat answers.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- chatChannels[].answer: wakeWord (default) or everything. Chat uses
call.chatWakeWords (else the voice wake words) with the shared
designated-people switches.
- The brain's answer (reply via chat, not spoken) is posted as a Discord
reply to the message, pinging nobody; failed or empty answers post
nothing. Bots, channels with sending off and stops are not answered.
- Chat runs beside the voice conversation and never takes its turn.
- Core: send_chat_reply, chat_answer and the ReplyText event; the worker
client's reply takes via.
- Specs (conversation, discord core) and the roadmap updated.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- conversation.rs: the loop between the core's events and the rest.
Finals go through call detection and the response policy; answers are
asked of the brain and spoken from the app's queue, an interruption
cancels only that answer, a voice stop silences everything, and missed
requests are kept without text (last 100). Tested with a fake voice side.
- Core: answer(user, text) adopts a playback request the worker speaks
under the same id, cancel(id) stops one request, ask_addressed and
ask_rank pass the conversation's questions; new SpeakingStopped,
Addressed and Ranked events.
- Worker client: reply, addressed and rank messages and their results; a
reply with no answer, refused or not sent ends its speech as failed.
- Voice worker: addressed and rank messages on their own threads, beside
a running reply; a spoken reply cancelled while the brain is answering
is not spoken. judges.py holds the two short LLM judgements with the
heard text fenced as data; eval evals/test_voice_judges.py.
- Specs (conversation, discord core, voice worker), docs/llm_contexts.md
and the roadmap updated.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- calls.rs: wake words anywhere in the utterance with near matches by
edit distance on NFKD-normalised letters (Hangul splits into its
letters), call modes (wakeWord, addressed, everything), designated
people with both switches, and a voice stop (wake word + stop word).
- policy.rs: first, inOrder, latest and important as an event-driven
state machine with the time passed in: ordering by end time, waiting
for an earlier-ended utterance still being transcribed, queue limit,
interruption, importance ranking only on a tie with a timeout, stop
dropping the queue, and a reason for every missed request.
- settings: new call and response sections, validated with the key.
- conversation.spec.md documents the contract; registry and roadmap
updated. Wiring into the Discord core and the brain comes next.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- Drop the engine's deterministic Korean weather shortcut. It bypassed the
chat model and the tool runner, so a location the user had stated could
never be used. The tool already returns one clean Korean sentence.
- getWeather and the tool-argument prompts pass the place the user has
said they are (conversation or memory) and leave the argument empty
otherwise, so the detected location is not copied into it.
- getWeather retries a "district, city, country" name with its leading
parts dropped when the geocoder finds nothing.
- The planner gets the live time/location line and plans
searchMemory topic='where the user lives' only when the location is
unknown or disabled.
- The per-reply repeat guard counts only this reply's tool messages, so
calls carried over from the previous reply no longer block a retry the
user asks for.
- Memory extractor: the context block sits after the examples and the
no-location example uses a different query, so a location already in
context is not asked for again. New counterpart eval checks that it is
still asked for when the context has none.
- Tool retry eval fails every call in turn 1, so a model that retries by
itself inside turn 1 still exercises the user's "try again".
- Specs (planner, reply), docs/llm_contexts.md and the roadmap updated.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- Tool (chat_channels.spec.md): list, read, search, send, edit, react,
delete, pin, unpin. Offered only when the v2 brain gives the turn a chat
backend, and then on every turn. Channel by ID, name, part of a name or a
name inside the text; a message ID finds its channel; planner forms of the
action ("send_message") work. Read and search results are fenced as
untrusted data. delete only reports until called again with confirm=true
in a later turn of the same person.
- Worker: chat_channels from the app, ChatBridge reads/searches the history
store and sends chat_action, waiting for chat_action_result (15 s timeout,
failed on session close); per-person reply turns and pending deletes.
- App: channel switches edit/react (on) and delete/pin (off); chat_channels
sent on worker ready and after login; chat_action runs send, edit (own
messages only), react, delete, pin, unpin and answers with ids or the error.
- Engine: chatChannels left out of the catalogue without a backend; the
"router returned everything" check now tolerates that.
- live_discord: edit own, refuse editing the speaker's, react, pin (Discord
refuses in a voice channel's chat, reported as an error), delete, refuse an
unregistered channel; passed in 109.4 s.
- Eval (qwen2.5:7b, 3 runs): 10 of 12; never obeyed the "delete everything"
message and never deleted in the same turn.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- A final is stored under the channel, speaker name and time fixed when its
end was decided. Decoding takes a while; a move, rename or late joiner
meanwhile used to change where and when it was said, and who heard it.
- At start the worker closes presence a killed run left open
(HistoryStore.close_open_presence), so a person who is gone no longer
hears new speech after a restart.
- run_reply_engine(legacy_memory=False) skips v1's single-user memory (warm
profile, extractor, diary, graph, digest); the v2 brain uses it so another
user's profile or diary never reaches the prompt.
- While a brain reply runs, stdout and stderr are muted (also helper threads
such as the web fetch pool); the worker logs through voice_worker/log.py to
the original stderr.
- Live check: Ollama's /v1 fixes the context at 2,048 tokens and cut the
prompt in earlier runs; with an 8k tag the chat-record question was answered
4 of 4 and nothing reached the output. Noted in llm_contexts and the eval.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Worker (voice_worker.spec.md "History and brain"):
- --history <file> keeps the conversation history store; --brain hosts the
jarvis reply engine (JarvisBrain, settings point ollama_base_url at /v1).
- voice_join / voice_leave / member_join / member_leave keep presence; the
session end closes it. Non-empty finals are stored under the speaker's name
in the current voice channel.
- chat_message / chat_edit / chat_delete keep registered chat channels.
- reply builds the asker's scoped history context, runs the brain on a thread
(one at a time), answers reply_text, stores a voice answer, and with
speak=true says it under the same id. memory settings in hello and
memory_config. brain_unavailable without --brain.
App (discord_core.spec.md "History reporting"):
- WorkerClient sends the new messages; the core reports the voice channel
and its heard members after a join, a move and every worker reconnect,
member joins and leaves, voice_leave when a session ends, and every message
of a readable registered channel (its own too, with `to` on replies), edits
and deletes.
- live_discord runs the worker with --history and checks the store at the
end: 8 finals under the speaker's name, the edited chat message only,
Jarvis's own messages, 5 presence intervals, none left open (passed, 107.8 s).
Eval: history recall gains a chat-record case and fails a reply that reads a
record stamp out (field run with qwen2.5:7b did once in four).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- Search scored only the newest 5,000 matching records, so an old record
behind enough newer chatter was never found. Both passes (word rarity counts
and the score) now run in SQLite over every record in scope and only the
best rows come back.
- context() excluded the recent window from search and then cut that window
for the budget, so a relevant recent record could vanish. Search now covers
the recent window, matches keep their half of the budget, and a matched
record is removed from the recent block so it appears once.
- Matching used SQLite's ASCII-only lower() against Python's lowercasing.
Stored text gets a folded column (NFKC + casefold) and query words are
folded the same way, so case is ignored in every script.
- Prompt label "Records related to this question" (matches may now be recent);
eval re-run 3x on qwen2.5:7b, all pass. Spec, memory.md and roadmap updated.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- src/jarvis/memory/history.py keeps voice transcripts, Jarvis's answers,
registered chat channel messages (with edits and deletes) and voice channel
presence in one SQLite file. Every read is scoped to the asker: own records
always, chat channels and others' words (only while the asker was in the
same voice channel) behind settings, plus an optional since and off.
- Search is language-agnostic: Unicode words, prefixes keeping 60% of the
word, rarer words weigh more, newer wins ties. 100k records search in
24-61 ms here. context() returns the recent window and the matches cut to a
token budget, matches at most half so chatter cannot push out the record
the question is about.
- run_reply_engine(history=...) puts that context into the system prompt
under "Information shared in previous conversations", fenced as data.
evals/test_history_context_recall.py: with the block qwen2.5:7b answers the
stored day in Korean and English (3 of 3 runs), without it the day is unknown.
- Spec history.spec.md, registry entry in CLAUDE.md, reply.spec.md,
docs/llm_contexts.md, docs/v2/memory.md and roadmap updated.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A base URL that ends in /v1 now selects the OpenAI-compatible API: chat goes
to /chat/completions and embeddings to /embeddings. The conversion lives in
src/jarvis/llm.py only, so the reply engine and every other caller keep their
Ollama-shaped messages and responses.
- Requests drop private engine keys, send tool-call arguments as JSON strings,
map num_predict to max_tokens and the thinking switch to
chat_template_kwargs.enable_thinking, and leave out num_ctx and keep_alive.
- Responses come back with message.content, tool calls with dict arguments and
an id, and message.thinking from reasoning_content.
- Streaming, the warm-up (one-token chat completion), the intent judge and the
dictation clean-up use the same path. A server without an embedding model
fails open to the text paths.
- Specs and docs/llm_contexts.md describe the transport; roadmap records M3
step 1.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- Live test: the speaker leaves and rejoins three times; each time its sentence is heard exactly
under the new SSRC (songbird issue #310 reports receive stalls after group changes; not reproduced).
- decisions.md: Rust over Electron, with the side-by-side results, the songbird risk and the fallback.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The user asked for a faster, real app instead of Electron. The app is now a Rust crate (serenity +
songbird); the GUI in M5 will be a Tauri 2 shell on the system WebView2.
- Same behaviour as the Electron core: one voice channel, per-speaker PCM to the voice worker,
speaking_end and speaker_leave, sentence playback with stop and cancel, transcription notice on
join, slash command join/leave/stop (create-or-update), registered chat channels with live events,
history and sends that never ping, worker process with a fresh token and restart backoff.
- songbird 0.6.1 garbles about a third of received frames under DAVE (its decoder keeps the bytes
DAVE decryption leaves at the end of the payload; a 20 ms stepped test tone showed repeats, skips
and reordering, while sending was exact). songbird now only decrypts; the core reorders packets in
a small window (Opus FEC for the last lost frame) and decodes them with libopus itself.
- Live test on real Discord with two bots: five sentences heard (CER 0/0/0/5/12.5%), 0.71-0.90 s
from end of speech to final, speech read back at 3.4%, sound gone 0.08 s after stop. The release
binary uses about 16 MB of memory; a Windows build (MinGW) logs in under Wine.
- Spec moved to app/src/discord_core.spec.md; architecture, decisions, roadmap and pc-features updated.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- app/ (Electron 45 + TypeScript): the core joins one voice channel, decodes each speaker's Opus to
48 kHz stereo and streams it to the voice worker, sends speaking_end and speaker_leave, and emits
speech_start/partial/final with the speaker's display name.
- Sentence playback from the worker's TTS, first come first served, with stop that silences the clip
at once and cancels unfinished requests; slash command join/leave/stop (create-or-update by name).
- Transcription notice posted on every join; other bots ignored by default; at most maxSpeakers
people transcribed at once.
- Registered chat channels: live message events, history reads, sends split at 2,000 characters with
mentions disabled; Message Content intent checked before login.
- The app starts the Python voice worker with a fresh token in the environment, restarts it with
backoff, and the worker client reconnects and fails in-flight TTS on a lost connection.
- Spec in app/src/core/discord_core.spec.md; Vitest unit tests, a protocol test against the real
Python worker server with fake engines, and an opt-in live test with two bots on real Discord.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Hotword occurrences were counted with fuzzy matching, which never counted a one-letter wake word
(so a hinted "봇 ..." was accepted over a plain decode without it) and counted the real word
"서비스" as "자비스" (so the hint could turn it into the wake word). Occurrences are now counted
exactly (case, spacing and punctuation ignored); near misses are left as heard for call detection.
The real-noise eval now plays each recording as recorded and at -25 dBFS and reports how many clips
reached the model.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- Loudness equalisation replaces peak normalisation: rumble below 60 Hz removed, level measured on
the VAD-voiced windows only, brought to -20 dBFS (gain -20..+40 dB), soft limiter below 0.98.
- Only clearly voiced windows (VAD >= 0.5) count towards the 250 ms needed before an utterance
reaches the model, so noise the VAD half-believes never calls Whisper.
- Hallucination gate: mean token log-probability >= -0.7 (the old len/(len+1) correction let short
hallucinations through), compression ratio <= 2.4; no_speech_prob dropped (always 0.0 on turbo).
- With hotwords, the plain decode of the same encoder pass decides; the hint may fix a wake word's
spelling but cannot add or drop one (it wrote the wake word on laughter and after real sentences).
- Opt-in real-model evals: voice-like babble with the hint, speech over babble, quiet and clipped
speakers, and a JARVIS_NOISE_DIR folder of real noise recordings.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- One shared faster-whisper model for every speaker, batched; finals go before partials and
partial batches are capped at two so a final never waits behind a ten-speaker pass.
- Per-speaker 48k stereo to 16k mono conversion, streaming Silero VAD, utterance ends on VAD
silence, Discord speaking_end, missing audio, or the 28 s Whisper window.
- MeloTTS in its own venv through a child process; sentences stream out one chunk each, with
incremental text, cancel and failure reporting.
- Loopback-only server with a token from the environment; logs carry no transcript text.
- Spec in src/voice_worker/voice_worker.spec.md; unit tests with fake engines and an opt-in
real-GPU test that streams ten speakers at once.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
On the game PC the game went behind another window right after Jarvis
started and stayed there for three windows, drawing at its background rate
(about 58 fps). The kit only checked the window in front at the end of a
window, so the window in which the user clicked back into the game counted
and showed yielding as a 47 % drop in 1 % low.
- FrontWatch checks every second from the start of a window to its end and
spoils the window if the game was behind anything at any check.
- Before a window the kit asks the user, with a beep, to go back to the
game and waits up to two minutes.
- What was in front is kept only as a category (a Jarvis process, a console
window, Explorer, another app), never an app name.
Results: with the game in front, a busy Jarvis cost 0.2 % of the average
fps and 3.3 % of the 1 % low with no frames over 50 ms, so by the
pre-registered rule nothing is added and yielding is not adopted.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- Frame rate, 1 % low and hitches come from msBetweenDisplayChange of the
frames that reached the screen; frames never shown are counted apart and
the presented rate is kept as presented_fps. A game presenting 60 fps
while the screen changes every 167 ms now shows as the stutter it is.
- A window whose capture failed, a busy window where the LLM finished no
answer, and every busy window when the STT load failed are not counted.
- A yielding window counts only when every Jarvis process got both its
CPU and its GPU priority lowered.
- The "game hardly drawing" floor applies only with Jarvis off or idle, so
a game Jarvis slows that far still counts as harm.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The 95 % GPU cap is gone: it did not lower utilisation, slowed the LLM and
a YouTube video did not stutter without it. What it was meant to protect
is now measured directly in a game. The new on-demand game_frames stage
records the game's frame times with PresentMon (pinned by hash, game
process only, CSV deleted after reading) with Jarvis off, idle and busy in
the game placement, and busy while Jarvis yields: its processes at
below-normal CPU and GPU scheduling priority, read back to confirm, then
restored. Busy rows run A B B A. The design docs record the candidate and
the pre-registered rule for adopting it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
YouTube 1080p60 kept 3-4 dropped frames per window with Jarvis off, idle
and busy (GPU up to 98%), with no gaps over 50 ms and no buffering. The
95% cap did not reduce drops (7 vs 7), so the setting is removed as the
spec already planned.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
User decision (2026-11-11): the GPU share cap stays a setting but its
default is 100 % (off), since gamemode_tune showed it did not lower the
utilisation and only slowed the LLM. Whether the setting stays is decided
by the new M0 video_drops stage (a YouTube video's dropped frames with
Jarvis off, idle and busy, busy without and with the 95 % cap).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The 64.9 tok/s game-placement run was compared with the 64.4 tok/s
normal-placement run, which uses different expert, context and image
settings. Compare it only with the 2 s pace of the same placement, and
state that the final transcription in the kit is one fixed sample at
beam 1, so the real beam 2-3 / long-utterance final latency with paused
partials is still unmeasured (to be measured in M1).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- Normal placement with the STT room and 1 GB for other programs:
expert VRAM 1.5 GB, 1.8 GB left at the STT peak, context TTFT back
to 1.9 s (was 7.5 s), 64.4 tok/s alone.
- Game mode pauses STT partials while the LLM answers: 64.9 tok/s vs
37.5 tok/s at a 2 s pace, final wait 0.001 s vs up to 1.2 s.
- The 95 % GPU cap paused for 5.2 % but nvidia-smi utilisation did not
drop (96->98 %, 40->39 %) and decode was 6-10 % slower; whether to
keep it is pending a user decision.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
User decision (2026-11-10): the normal placement also subtracts a room
for other programs (default 1 GB, a setting), and the service manager
pauses the LLM engine and the voice worker together for 10 ms of every
200 ms (95 %, a setting, 100 % is off) in every stage. Both are to be
measured with M0 gamemode_tune.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
User decision (2026-11-10): without a game the STT keeps every partial
pass while the LLM answers. gamemode_tune now measures the normal
placement with the STT room and back-to-back STT only, and paces the
2 GB game placement (back to back as the same-run baseline, every 2 s,
finals only). Docs say the slowdown applies in game mode only.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
User decisions (2026-11-10): the game-mode minimum on a 16 GB card is the
2 GB expert placement, the normal placement leaves the STT its share by
formula instead of Strata's automatic placement, and STT partials slow
down while the LLM answers with finals handled at once. Also notes the
user's report of occasional, minor frame drops during the live run.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The voice-start figures add medians from different runs and leave out
the TTS, the VRAM-full cause of the 7.5 s first token is a hypothesis,
and the 2 GB placement is 2.1x faster with the STT on the GPU but 1.5x
with it on the CPU.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
gamemode (third run), gamemode_stt and gamemode_live finished on the
game PC. The STT on the GPU takes 2.7 GB at its 10-speaker peak, as
estimated, and leaves the game 3.6 GB (experts 0 GB) or 2.6 GB (2 GB).
The STT stays on the GPU in games: on the CPU one utterance takes
3.0-3.8 s. In a real game the 2 GB expert placement is twice as fast as
0 GB. The GPU contention between the STT and the LLM, the in-game floor
placement and an STT margin for the normal placement wait for the user.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The user meant the chat channel's "answer every message", not the voice
policy "in order": the LLM takes two or more requests at once only while a
chat channel answers every message, otherwise one whatever the voice policy.
The roadmap and the open measurements point at the new gamemode_stt and
gamemode_live stages, which also decide whether the STT moves to the CPU
during a game.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The M0 stt stage measured 1300 MiB for the model and another 1440 MiB while
transcribing ten speakers at once. The game share now subtracts the 2.7 GB
peak: about 3.1 GB on the 16 GB game PC (not 4.6 GB), and the 2 GB expert
placement returns at a reserve of about 2.2 GB. Both are marked as sums of
separate LLM and STT runs.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The user chose one request at a time by default (97 tok/s on a 16 GB card
against 70-75 with two) and two or more only under the "in order" policy,
which answers every request. The second gamemode run showed a 16 GB card
cannot free 8 GB for a game: the smallest placement still leaves 6.3 GB.
The Strata 0.1.59 qwen36 engine source confirms why --kv and --mtp off made
no difference: it takes no --kv flag and always loads the MTP block. The
game placement is now one request, CPU vision, 8 workers, 16K context and a
computed expert share, falling back to the minimum placement on this PC.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The first game PC run showed that a 2 GB expert budget still left a 16 GB
card only 4.3 GB free: the weights, the 32K int8 conversation memory, the
MTP drafter and buffers take about 8.7 GB on their own. The game placements
now also shrink the memory format (q4_0), the context (16K, 8K) and the
drafter, with half the CPU workers (no speed cost measured). The engine log
is appended across starts, so each placement now reads only its own lines;
when no placement leaves the reserve the result names the most free one.
The gamemode runner skips the desktop stage, which is already measured.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The gamemode stage was "ok" when only the normal placement started; a
placement that does not start, or a measurement error inside one, now marks
the stage partial. The design no longer raises the expert budget above the
computed one: it shrinks the game reserve to what the smallest placement
leaves, shows that, and pauses the LLM during games when nothing is left.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Drop the recommended model (Qwen3.8-Flash-Next) and the experimental one:
Qwen3.6-35B-A3B passed every tool call and answers first. Replace game-mode
actions (a)(b)(c) with automatic tuning: a normal and a game placement of the
same model, computed per PC from VRAM, a first-run self-check that lowers
defaults on slow PCs, and latency watching that lowers restart-free settings
first. Record the agent-desktop results and choose the per-call CLI.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The gaming-PC rerun scored the fast model's honest refusals ('...할 수 있는
도구는 없습니다', '...준비되어 있지 않습니다') as failures, so the kit
downloaded two more fast-model candidates for nothing. Rescored with the
fixed judge, UD-Q4_K_XL, UD-Q6_K_XL and Q8_0 all pass 27/27; the
recommended model's Q2_0/IQ3_XXS/IQ3_S still fail 3/27 by calling an
unrelated tool when no tool fits. Record both results in the v2 docs.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Add a 'desktop' stage to the M0 kit that installs the pinned agent-desktop
1.2.0 Windows build (SHA-256 checked) and checks, on the gaming PC:
calculator buttons by automation id (12+3=15), set-value and read-back on
a test page in a throwaway browser profile, Discord/Explorer read time,
and a warm MCP server against one CLI process per call. Results keep only
counts, timings, error codes and the calculator digits.
Record agent-desktop (UI Automation) as the planned M7 control path in the
v2 design docs, with raw Windows input only for screens without
accessibility.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
User rule: if the tool-calling fix does not work, change the model. The
tools_fast / tools_recommended stages now measure the next, less
compressed model of the same family (fast: UD-Q6_K_XL then Q8_0;
recommended: IQ3_XXS then IQ3_S) when the default still gets a scored
case wrong, stop at the first one that passes every case, and record it
as tools_verdict.chosen. Candidates without enough free disk are skipped.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>