- ggml: read-only mmap module (unix-gated) + two-region Context (mapped weights / dirty caches) + segmented Metal buffer binding; llama loads GGUF weights as file-backed clean pages (jetsam-exempt) with owned-arena fallback (MAKEPAD_LLAMA_NO_MMAP=1). Route app dirty footprint 12GB -> 4-7GB; model load becomes lazy page-in; A/B byte-identical on 4B + 9B. - llama: fix graph-cache keying corruption — a graph keyed wider than the KV cache corrupted attention for any prefill batch >= 2 (flash op reads permute-node dims baked at build; view reconfigure never reached the kernel; masks were written cache-narrow). Masks now always fill the full graph key width and graphs key by 1024-buckets; reconfigure path removed. Verified byte-exact vs per-length reference across batch 1/2/8/64/512, short+long prompts, mmap on/off, plus a two-session concurrency probe. - voice: passive VoiceWaves no longer register the global audio-input callback (the invisible caption-bar wave stole mic audio — last registrant wins — and spawned duplicate whisper workers); whisper back to F16 default (voice Metal library has no quantized kernels; q5_0 failed every GPU matmul); raw transcripts render immediately. - route: kokoro TTS voice output (speaker toggle; streams reply sentences, announces nav maneuvers + arrival) with barge-in — voice activity on the mic stops playback instantly; dispatcher context 8k -> 32k (hybrid KV is 12/48 layers, ~48KB/token); window caption bar suppressed under studio. - llama-generate: --max-context/--prefill-batch-size + state fingerprints; new llama_concurrent_probe bin. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
25 lines
1.3 KiB
TOML
25 lines
1.3 KiB
TOML
[package]
|
|
name = "makepad-app-route"
|
|
version = "0.1.0"
|
|
edition = "2021"
|
|
description = "AI-driven voice route/trip planner on MapView (see route.md)"
|
|
license = "MIT OR Apache-2.0"
|
|
|
|
[dependencies]
|
|
makepad-widgets = { path = "../../widgets", version = "2.0.0", features = ["maps", "voice"] }
|
|
makepad-map-nav = { path = "../../libs/map_nav" } # M1: route.* tools (search + A* graph)
|
|
makepad-tts = { path = "../../libs/tts" } # kokoro voice output (🔊)
|
|
makepad-geodata = { path = "../../libs/geodata" } # M1: geo.* corridor queries, rain radar
|
|
makepad-mbtile-reader = { path = "../../libs/mbtile_reader" } # terrain drape landcover source
|
|
makepad-ai = { path = "../../libs/makepad_ai" } # M1: cloud agent + tool definitions
|
|
serde_json = "1" # geodata FeatureHit attrs
|
|
|
|
makepad-llama = { path = "../../libs/llama" } # M2: local dispatcher LLM (pure-Rust ggml)
|
|
|
|
makepad-converse = { path = "../../libs/converse", features = ["local-llm"] } # M3: attention-gate filter
|
|
|
|
# Planned wiring (see route.md milestones) — enable as each phase lands:
|
|
# makepad-tesla = { path = "../../libs/tesla" } # M5: vehicle SoC / charge state
|
|
|
|
[dev-dependencies]
|
|
makepad-test = { path = "../../libs/makepad_test", version = "0.1.0" }
|