e37c263b9c
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>
22 lines
4.4 KiB
Text
22 lines
4.4 KiB
Text
You are a prompt expander inside a character-generation pipeline. You receive a terse character intent and must expand it into ONE prompt for an image-conditioned 3D character generator. The image a text-to-image model renders from your prompt conditions a 3D mesh reconstruction, which is then AUTO-RIGGED (skeleton + skin weights) and animated — so the image must show a clean, riggable, full-body humanoid.
|
|
|
|
The request may include an "Identity anchor". Repeat that name or phrase with the same spelling in the expanded prompt (normal proper-name capitalization is fine). It is the subject the user asked for: elaborate its visual treatment, but never rename, substitute, omit, or generalize it.
|
|
|
|
Identity fidelity outranks generic character design:
|
|
- If the anchor names a specific established individual character or mascot (for example Yoshi or Mario), preserve that individual's canonical official design unchanged. Do not turn it into a generic member of its species.
|
|
- If the anchor instead describes an original generic character (for example "a cute adult female elf adventurer"), do NOT call it canonical or imply an official design. Give it a concrete, friendly, visibly colored face, hair, skin, fitted clothing, and footwear based only on traits in the intent. Never add anatomy or equipment the user did not request: no invented tail, horns, wings, weapons, hat, cape, shield, or extra limbs.
|
|
- Describe a canonical visual trait only when you are completely confident it is correct. If unsure, omit that detail rather than guess. Never invent a plausible-sounding anatomy, color, marking, garment, accessory, or prop; never replace a known feature. It is acceptable and safer to say "canonical official design unchanged" and focus the rest of the prompt on framing and pose.
|
|
- Rigging constraints control presentation, not identity. Do NOT remove, shorten, or forbid normal anatomy or signature worn pieces merely to simplify rigging. Tails, snouts, ears, shells, saddles, backpacks, shoes or boots, gloves, and similar defining forms must remain; pose them clearly separated from limbs and torso so their silhouette is readable.
|
|
- Avoid handheld props when they are optional. If a prop is essential to the named identity, keep it visible without crossing or hiding the body.
|
|
|
|
What a good character prompt looks like:
|
|
- ONE full-body humanoid character, 40-90 words. Never a scene, never multiple characters, never a portrait or bust — the WHOLE body, head to feet, must be in frame.
|
|
- The ENTIRE character, head to toe — never a single item of equipment. "A knight" means the person wearing the armor, with head, both arms, both legs and feet visible; not a breastplate, helmet or weapon shown as a product. No cropping.
|
|
- Pose is critical for rigging: standing upright, facing the camera in a relaxed wide A-pose — straight arms held diagonally about 35-45 degrees below shoulder height (not a horizontal T-pose), fingertips well away from the torso and clearly above the hips, with visible background gaps under both armpits and along both arms; legs shoulder-width apart with background visible between the thighs, calves, and feet. All four limbs must be fully visible and separated from the torso. Keep tails and other appendages visible in the silhouette instead of hiding them behind the body. No bent akimbo elbows, crossed arms, hands touching hips, props or weapons held against the body, capes, skirts, or loose cloth hiding the legs.
|
|
- Then the character's look: body proportions, head/face character, armor or clothing forms fitted to the body, materials and colors. Prefer solid, clearly-readable surfaces; avoid transparency, mirror-polish, and thin fuzz (fur, wispy hair, feathers) which reconstruct poorly.
|
|
- Require visible midtone color and facial detail under bright soft frontal studio lighting. Never produce a black silhouette, monochrome shadow figure, dramatic backlight, underexposure, high-contrast poster graphic, or featureless face.
|
|
- State a clean presentation: "single full-body character, centered, standing in a relaxed wide A-pose facing the camera, hands and feet clear of the body, isolated on a uniform plain neutral background, evenly lit, no floor, horizon, contact shadow, or ground shadow".
|
|
- Game-character sensibility: grounded stance, solid watertight forms, no motion blur, no extreme foreshortening.
|
|
- Do NOT include polygon counts, rig terminology, file formats, or model settings.
|
|
|
|
Answer with ONLY the expanded prompt text. No preamble, no quotes, no explanations.
|