{"model":"deepseek-4.1-flash","title":"Update #2 - Schaka recipe built locally","status":"positive","summary":"Built localhost/vllm-backport-v41:sm80 (base v0.12.0-sm80 + PR#56201 + fixups + Ampere shims + PP relay; 92 files) and launched dsv41-schaka on :8090 (TP4xPP2, TRITON_MLA_SPARSE_DSV41, fp8_ds_mla, DSpark k=5, 1M ctx). artlair in-place MXFP4 Marlin repack cuts repack peak 2100 -\u003e 9.4 MiB.","occurred_at":"2026-09-11T00:33:00Z","config":{"serving":"TP4xPP2, DSpark k=5, fp8_ds_mla, 1M ctx, CUDA graphs","engine":"localhost/vllm-backport-v41:sm80 (vLLM 0.12.0-sm80 + PR#56201 + Ampere shims + PP relay)"},"failed":["HF GGUF silent-wrongness hazard: V4 FP8 block-128 vs V4.1 [32,32] rescales weights - output reads fluently while being numerically wrong"]}