{"model":"deepseek-4.1-flash","title":"PP relay correctness fix adopted - prefill triples","status":"positive","summary":"Adopted Schaka's relay fix (topk_indices on the hop + index_k_cache guard): TP1xPP6 prefill 5.6k @32k / 7.2k @128k. TP4xPP2 remains decode-optimal (512k c64 129.6, correct 12/12). TP2xPP4 / TP1xPP8 still boot but needle 0/3 (relay cannot handle TP splits) - rejected.","occurred_at":"2026-09-11T15:09:00Z","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)"},"benchmarks":[{"metric":"prefill_tok_s","value":5600,"unit":"tok/s","context":"TP1xPP6 @32k (7.2k @128k) - ~2.5-3x TP4xPP2"}],"failed":["TP2xPP4 and TP1xPP8: needle 0/3 after relay fix - arbitrary kv-group TP splits still unsupported"]}