Spaces:
Paused
Paused
Commit History
-
deploy thoxfast 058d0e08b664
Verified signatureDeploy the production-validated provider payload from ThoxRoute main 058d0e08b66487267a9f9389e6d0679f1b2361a7 (PR #69). Preserves existing .gitattributes.f73e85a -
Give fast model full CPU quota
Verified signature45b2646 -
deploy: warm shared fast-chat prefix
Verified signatureExact provider tree from ttracx/thoxroute main c19c9b55d9b8bd33a32be5130697c0a019afb3c8; local CI 874 tests and Q4_K_M container validation passed.bf13452 -
Restore Q4_K_M after fidelity failure
Verified signatureExact source: ttracx/thoxroute@21f69dda9b31838ffa69de7c61de9565d7fc69575928492 -
Use lower-latency interactive Q4_0
Verified signatureExact source: ttracx/thoxroute@cf9bdd16421e23a5f11fc363d420cf0344b573ade15b3f8 -
Restore two-context interactive topology
Verified signatureExact source: ttracx/thoxroute@49d114b14ba5f1a43fcb93d2f5a2b86a2e6e5e0c2a84032 -
Keep interactive streams live during inference
Verified signatureExact source: ttracx/thoxroute@f56b047f0c269ed6559d728ad42e06cd8bb5420145f3b9d -
Bound interactive inference backpressure
Verified signatureExact source: ttracx/thoxroute@4f71bbe0dea6d8f029c06a66f010928f0f3635504dda59b -
warm interactive contexts before readiness
Verified signatureExact source: ttracx/thoxroute@e38f1311bc9ce7f88df471c20924f244a500a470c8944a3 -
deploy latency-bounded interactive pool
Verified signatureExact source: ttracx/thoxroute@755605a45facd12e12c947d4cab7b886d80bd218955b771
-
THOX Rust Coder: Qwen3-Coder-REAP-25B-A3B-Rust GGUF via conda-forge llama.cpp
Verified signaturea0ef6f6 -
initial commit
Verified signature2fff547