makepad/libs/asset/chat/tests/live_fleet_probe.rs
Admin e37c263b9c ai backbone: the hub era — makepad_ai deleted, every backend is a hub pipe; machine residency elections, job leases, ETA placement; the store only stores; creator pipelines run in the app (aicore)
Squashed from work:
- asset-ai: FastH3 4-step fast video backend; clip keyframes on the wire
- asset-ui: loop video chains — text→image→video that ends where it began
- h3: safetensors -> pruned-Q4_K GGUF quantizer for the 24GB DiT tiers
- h3_quant_gguf verify: row-error gates calibrated to the measured Q4_K floor
- asset-ai realtime: the feedback loop — the source anchors, the drifted frame inits
- asset-ai realtime: a feedback loop survives a resize and travels by default
- asset-ai realtime: the feedback loop frees itself from the feed handshake and pauses for its listener
- asset-ai realtime: the outbound encode leaves the loop's critical path
- asset-ai ocr: the ocr domain — Chandra 2 at page resolution, and the tower goes planner-owned
- llm slots: a lane can hold an image span — embedding prefill and a rope cursor of its own
- vision tower on CUDA: the encode leg gets its two missing kernels
- llm/ocr: one M-RoPE grid encoder for both image paths, and a livelock made an error
- vision tower on CUDA: the f16 GEMM keeps the precision it was throwing away
- live: a feed that moves box takes its trip with it — one seed image
- vision tower on CUDA: the tiled attention becomes bit-exact, and tensor cores go
- llm prefill on CUDA: the MMA attention kernel gets the tile a 4-to-1 model needs
- asset-ai ocr: the CUDA encode lane joins the integration — vision-parity sits beside run's three arms, and the kernels
- Merge branch 'ocr-perf-integration' into work
- asset-ai: the live anchor can follow the trip, and text leaves the 5090
- asset-ai: the camera moves the world, and the world starts still
- asset-import: the EA strategy classics, in the one 2D contract
- rtsmap: one seeded generator for tiled strategy maps
- asset-ui: one card for the strategy classics, with a pack dropdown
- asset-ai: music3 reference-audio path, ocr/h3 backends, registry
- asset: mp4 sample index for range-streaming, chat tools, import profiles
- cnc: tiberium is twelve growth frames, not twelve empty variants
- platform: native file and save dialogs, in-house on all three desktops
- chat: the scan holds out for a lane home
- chat: a full home queues you — take the free lane
- chat: the preload has a percentage, and the boundless cap stops showing
- llm cuda: the 32x2 attention tile — even GQA ratios stay on MMA
- sa3 gets a bake path: the sfx model's tables precomputed by a diffusion-side bin
- sqlite_query: anti-join regression test
- td import: HARV's second frame block is its harvesting cycle, not a turret
- asset-ui: sprite enhancement runs on the 32B dev DiT — distillation, not the prompt, was the ceiling
- ai-hub: makepad-asset-ai becomes makepad-ai-hub at libs/ai/hub, the chat pane becomes makepad-chat-ui, the service bin
- asset-ui: test health fixtures grow the realtime field they were born without
- ai-hub: one home at ~/.makepad — weights/ run/ cache/ logs/, the service cache migrates from ai_content by a single re
- ai-hub: subprocess workers die with the node — process groups everywhere, PDEATHSIG on linux, one KILL_ON_JOB_CLOSE Jo
- ai-hub: the hub object — AiHub::in_process, pipes vocabulary, and the local LLM engine generalized out of mpfiles (aic
- strict-json: the dependency-free JSON module gets its own crate; asset-client re-exports it so nothing downstream move
- ai-hub: the machine layer — node entries, the 0600 machine token, and the residency election that IS the lock (aicore
- ai-hub: MPHUB1 — the fabric beacon only dedicated nodes can send (aicore §4)
- ai-hub: job leases — work lives only while it is renewed (aicore §8)
- asset-creator: the pipeline library is born — specs, the deps gate, and the derived-state law (aicore §9)
- ai-hub: RAM residency facts — the CPU-side twin of residency.rs (aicore §3)
- ai-hub: ETA placement primitives — relative GPU throughput, the four-term estimate, and an observable breakdown (aicor
- ai-hub: leases go live on the wire — origin fields on submit, /job/<id>/keepalive, /bye, and the reaper that cancels w
- ai-hub: the chat providers move in — fleet qwen, openai, grok, claude/codex/grok CLIs, the responses driver, and the w
- asset-creator: the engine — one pipeline run against the hub, deps-gated, spliced, cancellable, resumable-by-construct
- ai-hub: the machine node mode — --machine binds loopback, registers in ~/.makepad/run, and exits on its own once idle
- asset-creator: makepad-creator-run — the detached client for runs that must outlive a window (aicore §9)
- ai-hub: a native Claude Messages-API provider — API-key or Claude Code OAuth, bounded SSE streaming, injected tools (a
- route + converse: off makepad_ai — the Agent seam moves to converse, route's cloud dispatcher rides the hub's Claude p
- asset-creator: the preset tables move in — fifteen chain-policy constants shared by every creator app (aicore §9 / P6)
- makepad_ai is deleted — every backend is a hub pipe, the agent seam lives with its consumers (aicore §14, decided 2026
- ai-hub: loads hold the machine residency election — set_model_state claims on Loaded and publishes the service port (a
- ai-hub: chats run the machine election — route to a serving holder, wait on a loading one, claim and publish when open
- ai-hub: pick_for_domain_eta — ETA-ranked placement over the shared hard-filter core (aicore §6 / P4)
- asset-creator: the engine picks a provider per stage at dispatch time — a chain's later stages see fresh fleet state (
- ai-hub: the fabric secret gates the service HTTP surface — bearer on everything but /health and the ticketed peer path
- vj: DREAM runs execute in the app — pipelines.rs becomes the run it used to watch (aicore §9 / F1)
- asset-creator: the runner — generate one thing and put it in the catalog, one implementation for every surface (aicore
- chat-ui: the session runs in the app — no broker anywhere on the chat path (aicore P8 / F5)
- asset-store: assets.query is a first-class query endpoint — the bounded SQL surface outlives the broker (aicore P8 / F
- asset-creator: CreatorTools — the chat tool pack for a store that only stores (aicore §9 / P8)
- asset-store: the shrink — the store stores (aicore P7)
- importer + asset-server host: the coordination era ends (aicore P7)
- store config purge + asset-ui goes fleet-direct; the derive protocol gets its route proof (aicore P7)
- client + chat dispatcher: the dead wire comes out (aicore P7/P8)
- ai-hub: 0.3.0 — the health version says which era a node runs
- ai-hub: the default fleet is 'gen' — apps hear the LAN without env plumbing
- ai-hub: the preload note percents the prefill, not the job bar
- ai-hub: conversations keep their KV — the wire mirror, the lane identity, the in-turn dynamic context (aicore §7)
- ai-hub: an open-think model is thinking from its first token
- libs: the zero-warning sweep — stitch casts say what they mean, xatlas keeps upstream's surface quietly
- zero-warning sweep, round two — the first full-workspace pass
- zero-warning sweep, round three — the model lanes and the deep examples
- zero-warning sweep, round four — the last stragglers
- zero-warning sweep, round five — vj and chat-ui
- zero-warning sweep, round six — three cascades

Co-authored-by: Claude <info@makepad.nl>
2026-09-01 16:46:31 +02:00

114 lines
4.8 KiB
Rust

//! What the serving box does with the generation session's ACTUAL system
//! text. Ignored by default; it needs a live fleet box.
//!
//! ```text
//! MAKEPAD_FLEET_BASE=http://10.0.0.165:8123 \
//! cargo test -p makepad-asset-chat --test live_fleet_probe -- --ignored --nocapture
//! ```
//!
//! It answers one question a model's self-report cannot: given the tool
//! table we render, does the box emit `image.generate` for "make me a
//! picture of a unicorn"?
use makepad_asset_chat::provider::{ChatProvider, ProviderEvent, TurnInput};
use makepad_asset_chat::qwen::{FleetQwenChatProvider, HttpFleetTransport};
use makepad_asset_chat::toolcall;
use makepad_asset_chat::tools;
use makepad_asset_chat::wire::{ChatMessage, ChatRole};
use std::time::{Duration, Instant};
fn run(label: &str, capabilities: &str) -> String {
run_with_history(label, capabilities, Vec::new())
}
fn run_with_history(
label: &str,
capabilities: &str,
history: Vec<ChatMessage>,
) -> String {
let base = std::env::var("MAKEPAD_FLEET_BASE").expect("MAKEPAD_FLEET_BASE");
let mut provider = FleetQwenChatProvider::new(HttpFleetTransport, vec![base]);
let system = toolcall::render_system(&tools::definitions(), capabilities);
let mut messages = history;
messages.push(ChatMessage::new(ChatRole::User, "make me a picture of a unicorn"));
let input = TurnInput {
system,
messages,
tools_enabled: true,
dynamic_context: String::new(),
};
provider.begin_turn(&input).expect("begin turn");
let deadline = Instant::now() + Duration::from_secs(180);
let mut text = String::new();
while Instant::now() < deadline {
for ev in provider.poll() {
match ev {
ProviderEvent::Delta(t) => text.push_str(&t),
ProviderEvent::Done { text: t } => {
let out = if t.is_empty() { text.clone() } else { t };
println!("=== {label} ===\n{out}\n");
return out;
}
ProviderEvent::Error(e) => panic!("{label}: provider error {e}"),
_ => {}
}
}
std::thread::sleep(Duration::from_millis(50));
}
panic!("{label}: timed out; partial={text}");
}
#[test]
#[ignore = "needs a live fleet box (MAKEPAD_FLEET_BASE)"]
fn the_box_calls_image_generate_from_our_tool_table() {
let out = run("bare table", "");
assert!(
out.contains("image.generate"),
"the box ignored the advertised tool table"
);
}
/// The failure the user hit: a worker-less server, a few turns of ordinary
/// chat first, and the model answers "I don't have an image-generation tool
/// available in this session" instead of calling the tool that is sitting in
/// its own list. The capability doc's wall of UNAVAILABLE is what did it —
/// so the generation profile now says, before that list, that the generate
/// tools are a different path and are live.
#[test]
#[ignore = "needs a live fleet box (MAKEPAD_FLEET_BASE)"]
fn an_offline_operation_worker_does_not_silence_the_generate_tools() {
// The real assembled context of a `gen` session on a server whose
// operation worker is down.
let dynamic = format!(
"LIVE STORE right now (kind count): mesh 4004, image 271, audio 245\n\n{}{}",
"The *.generate tools (image, video, audio, speech, music, mesh, world, \
character) run on the CONNECTED APP's fleet and are AVAILABLE in this \
session. The registered operations below are a separate, optional path \
for deriving from an existing asset; their worker status never applies \
to the generate tools.\n",
"Registered operations (for operation.create):\n\
- mesh.from_image.v1 [UNAVAILABLE: the worker is currently offline]: image to mesh\n\
- depth.from_image.v1 [UNAVAILABLE: the worker is currently offline]: image to depth\n\
Operations are created in namespace 'gen'. Inputs must be exact revisions \
bound to this session.\n",
);
let caps = makepad_asset_chat::context::assemble(
makepad_asset_chat::context::ClientProfile::Gen,
&dynamic,
);
// The same warm-up the user's window had before it refused.
let history = vec![
ChatMessage::new(ChatRole::User, "hi qwen"),
ChatMessage::new(ChatRole::Assistant, "Hi! I'm Qwen. How can I help you today?"),
ChatMessage::new(ChatRole::User, "test"),
ChatMessage::new(
ChatRole::Assistant,
"All good — I'm here and responsive. What can I do for you?",
),
];
let out = run_with_history("worker offline, warmed-up chat", &caps, history);
assert!(
out.contains("image.generate"),
"an offline operation worker talked the model out of generating"
);
}