Prompt 3.0.4: impulsive broad search rule

Rebuild backend bundle: plan-finder tools + billing routes

Ship paced-reveal fix bundle

nudge runtime

Ship smoother v2 (paced tail drain) bundle

rebuild trigger

diag: route-entered marker

diag: surface socket errors

diag: dump raw frames

diag: temp live-frames route (frame timings only, no key exposure)

backend: frontend bundle with constant-tick stream smoother

backend: streaming emitter + phrase-table sync (~1.5s TTFT)

backend: rebuild with api/search rename

search: drop bing from the race — it wins with bot-degraded junk

backend: ship race-enabled standalone as respite-backend.tar.xz (runtime contract)

search: cap instant-answer wait at 1.5s so the race latency wins

search: bundle Next backend with SearXNG engine race (linux-only natives)

Fix backend bundle: include standalone .next (BUILD_ID was dropped by the glob copy)

Rebuild backend bundle: theme fix + SearXNG search

Rebuild backend bundle with SearXNG-backed search route

Bootstrap searxng venv without pip + get-pip fallback (slim image lacks ensurepip)

Recreate stale searxng venv when its python predates 3.11

Pick newest python for the searxng venv (master needs 3.11+; base has 3.10)

Surface searxng bootstrap errors in status route

Add SearXNG userspace runtime with DDG + Google engines

CORS on the proxy so the Vercel frontend can call the space directly

Clean diagnostics: keep status/version/proxy only

childenv: include value lengths

Add child-env diagnostic

Add envcheck diagnostic (key names only)

Proxy: type the Request param for FastAPI DI

Proxy: forward method/query/body/headers, stream SSE both ways

Ship Next.js standalone backend bundle (6MB xz) for userspace Node runtime

Add Node runtime capability probe (userspace, no root)

Expose live audio generation API route

Pass Qwen prompt through async job

Fix Qwen multimodal prompt content

Diagnose Qwen generation termination

Enable Qwen reasoning by default

Add torchvision for Qwen processor

Add CPU Qwen3.8 inference job

Normalize TinyLlama chat template inputs

Fix CPU TinyLlama PyTorch initialization

Accept JSON body for TinyLlama inference

Restore Gradio ZeroGPU Space metadata

Add Transformers runtime for TinyLlama

Add CPU TinyLlama inference to Gradio Space

Remove incompatible Docker entrypoint

Switch Space to Docker experiment

Add standalone CPU server entrypoint