Spaces:
Running on Zero
Running on Zero
Commit History
-
fix(gate): accept x-ael-key alongside x-respite-key; drop dead key read
piThe Respite -> Ael rename renamed the app's copy of the shared-key header to X-Ael-Key while this gate kept reading x-respite-key. Those are different header *names*, so every protected request was rejected with 401 from a perfectly correct key — search and audio, on every deployment, silently. Header names are case-insensitive, so the rename looked harmless in review; it was not. c9fb4ed already restarted the Space so SEARX_API_KEY is loaded, which is why the only remaining fault was the header name. - Accept either spelling, x-respite-key first as canonical. The app now sends the canonical one; the alias keeps any build in flight during a rename from 401ing, and must not be dropped without checking deployed builds. - Drop the module-level _SPACE_KEY read of "RESpite_SPACE_KEY" — a case-mangled secret name that never existed and always read "". It was masked because the middleware rebinds the same name per request, so the broken read was dead code one deleted line away from…b3596ff
-
security: give the searx bridge its own gate key
RESPIRE_SPACE_KEY guards the audio and inference routes; the searx bridge now checks SEARX_API_KEY instead, so search access can be rotated without touching the audio path. Both fail closed when unset. 🤖 Generated with Codebuff Co-Authored-By: Codebuff <noreply@codebuff.com>64e2bf0 -
security: key-gate the searx bridge
The shared-secret gate covered /respite/search but the searx bridge mounts under /respite/v2/searx/, which the prefix list missed — the search endpoint answered unauthenticated requests. Same fix shape as the audio gate: same compute, same key. 🤖 Generated with Codebuff Co-Authored-By: Codebuff <noreply@codebuff.com>d2bc83a
-
chore: remove the temporary env diagnostic
Verified signature8251b07 -
chore: temporary names-only env diagnostic
Verified signature977bce0 -
fix: read the shared secret per request; add a names-only env diagnostic
Verified signature3fee5c9 -
chore: rebuild so the container picks up RESPIRE_SPACE_KEY
Verified signaturebb7d60b -
security: require a server-held shared secret for the compute routes
Verified signature13dcd95 -
security: overwrite Gradio's reflected CORS on /respite/*
Verified signaturef1ffefe
-
Add SearXNG userspace runtime with DDG + Google engines
BuffyBootstrap pattern mirrors node_runtime: fetch source tarball, pip-install into a userspace venv on first boot, launch bound to 127.0.0.1:8888. FastAPI bridge exposes /respite/v2/searx/search (JSON API proxy).42f6abb -
Add Node runtime capability probe (userspace, no root)
Buffy- node_runtime.py: downloads official Node tarball to local cache, starts the Next.js standalone backend on 127.0.0.1:3210 as a child process - node_probe.py: /respite/v2/node/{status,version,server,proxy} routes to prove bootstrap, child exec, localhost HTTP, and reverse-proxying - app.py: wire probe routes into the monkey-patched create_app - drop committed __pycache__ noiseba9ed6e
-
Expose live audio generation API route
Verified signature58e1c4d -
Pass Qwen prompt through async job
Verified signatured3842e2 -
Fix Qwen multimodal prompt content
Verified signature024ac84 -
Diagnose Qwen generation termination
Verified signature8f2c140
-
Enable Qwen reasoning by default
Verified signatureca39a8e
-
Add CPU Qwen3.8 inference job
Verified signaturee8b6e3f -
Normalize TinyLlama chat template inputs
Verified signature9a9fa33 -
Fix CPU TinyLlama PyTorch initialization
Verified signature4380259 -
Accept JSON body for TinyLlama inference
Verified signaturefc0acf8 -
Add CPU TinyLlama inference to Gradio Space
Verified signature95167c1 -
Test RAM loaded GGUF without pipe deadlock
Verified signature3956ce2 -
Rebuild llama.cpp with stable CPU backend
Verified signaturef3c6ec5 -
Limit CPU thread pools to cgroup CPUs
Verified signature1b0a401 -
Prevent llama output pipe deadlock
Verified signaturea8a716e -
Probe llama CPU process state
Verified signature8ef48d6 -
Run loader diagnosis as background job
Verified signature3fdbf36 -
Add CPU model loader diagnostics
Verified signaturea5de9d9 -
Run CPU benchmarks asynchronously
Verified signature318e7cb -
Diagnose single-thread CPU model loading
Verified signature50b3226 -
Add tiny GGUF CPU control test
Verified signature74b57c5 -
Use RAM loaded CPU model benchmark
Verified signature6a6f8c0 -
Separate model loading from CPU warmup
Verified signature005dc00 -
Add CPU model smoke diagnostics
Verified signaturea903372 -
Serialize CPU benchmark requests
Verified signature75c0e18 -
Keep local model benchmarks CPU only
Verified signature677b59e -
Select llama benchmark executable
Verified signature0329f82 -
Fix llama benchmark binary cache
Verified signature6e550c2 -
Use llama bench for bounded CPU measurements
Verified signature4c94260 -
Fix CPU benchmark diagnostics
Verified signaturece1fcf1
-
Restore stable Space server launch
Verified signaturebc0692a -
Add share=True to bypass broken proxy
Verified signature15d9481 -
Restore working create_app monkey-patch with staticmethod
Verified signatureafb0c33 -
Background thread inject routes into running Gradio server
Verified signaturec7f0d89 -
Fix: inject routes via background thread after launch
Verified signature5e3fbbf -
Fix: patch App.__init__ to inject routes into the actual served app
Verified signatured0e8976 -
Fix: monkey-patch without staticmethod wrapper, spaces before torch, ssr_mode=False
Verified signaturee244295 -
Fix route injection: try demo.app directly, fallback to monkey-patch
Verified signaturee3186fc -
Add model filter param to benchmark endpoint
Verified signaturea76c80d -
Fix: build llama.cpp with static linking, find binary after build
Verified signature2ab0e84