Commit History

  1. Clean diagnostics: keep status/version/proxy only

    Buffy
    6da16aa
  2. childenv: include value lengths

    Buffy
    fcbf608
  3. Add child-env diagnostic

    Buffy
    79b68e7
  4. Add envcheck diagnostic (key names only)

    Buffy
    e59da61
  5. Proxy: type the Request param for FastAPI DI

    Buffy
    bb11950
  6. Proxy: forward method/query/body/headers, stream SSE both ways

    Buffy
    abb3d7d
  7. Ship Next.js standalone backend bundle (6MB xz) for userspace Node runtime⁠

    Buffy
    71ea0d0
  8. Add Node runtime capability probe (userspace, no root)⁠

    Buffy
    ba9ed6e

  1. Expose live audio generation API route

    Verified signature
    58e1c4d
  2. Pass Qwen prompt through async job

    Verified signature
    d3842e2
  3. Fix Qwen multimodal prompt content

    Verified signature
    024ac84
  4. Diagnose Qwen generation termination

    Verified signature
    8f2c140

  1. Enable Qwen reasoning by default

    Verified signature
    ca39a8e

  1. Add torchvision for Qwen processor

    Verified signature
    5be5b50
  2. Add CPU Qwen3.8 inference job

    Verified signature
    e8b6e3f
  3. Normalize TinyLlama chat template inputs

    Verified signature
    9a9fa33
  4. Fix CPU TinyLlama PyTorch initialization

    Verified signature
    4380259
  5. Accept JSON body for TinyLlama inference

    Verified signature
    fc0acf8
  6. Restore Gradio ZeroGPU Space metadata

    Verified signature
    1d49bf6
  7. Add Transformers runtime for TinyLlama

    Verified signature
    feb9a9b
  8. Add CPU TinyLlama inference to Gradio Space

    Verified signature
    95167c1
  9. Remove incompatible Docker entrypoint

    Verified signature
    ce7edba
  10. Switch Space to Docker experiment

    Verified signature
    9c40af1
  11. Add standalone CPU server entrypoint

    Verified signature
    707d14d
  12. Test standalone llama server container

    Verified signature
    d55cd28
  13. Test RAM loaded GGUF without pipe deadlock

    Verified signature
    3956ce2
  14. Rebuild llama.cpp with stable CPU backend

    Verified signature
    f3c6ec5
  15. Limit CPU thread pools to cgroup CPUs

    Verified signature
    1b0a401
  16. Prevent llama output pipe deadlock

    Verified signature
    a8a716e
  17. Probe llama CPU process state

    Verified signature
    8ef48d6
  18. Run loader diagnosis as background job

    Verified signature
    3fdbf36
  19. Add CPU model loader diagnostics

    Verified signature
    a5de9d9
  20. Run CPU benchmarks asynchronously

    Verified signature
    318e7cb
  21. Diagnose single-thread CPU model loading

    Verified signature
    50b3226
  22. Add tiny GGUF CPU control test

    Verified signature
    74b57c5
  23. Use RAM loaded CPU model benchmark

    Verified signature
    6a6f8c0
  24. Separate model loading from CPU warmup

    Verified signature
    005dc00
  25. Add CPU model smoke diagnostics

    Verified signature
    a903372
  26. Serialize CPU benchmark requests

    Verified signature
    75c0e18
  27. Keep local model benchmarks CPU only

    Verified signature
    677b59e
  28. Select llama benchmark executable

    Verified signature
    0329f82
  29. Fix llama benchmark binary cache

    Verified signature
    6e550c2
  30. Use llama bench for bounded CPU measurements

    Verified signature
    4c94260
  31. Fix CPU benchmark diagnostics

    Verified signature
    ce1fcf1

  1. Restore stable Space server launch

    Verified signature
    bc0692a
  2. Add share=True to bypass broken proxy

    Verified signature
    15d9481
  3. Restore working create_app monkey-patch with staticmethod

    Verified signature
    afb0c33
  4. Background thread inject routes into running Gradio server

    Verified signature
    c7f0d89
  5. Fix: inject routes via background thread after launch

    Verified signature
    5e3fbbf
  6. Fix: patch App.__init__ to inject routes into the actual served app

    Verified signature
    d0e8976