ent 9b7be16bee int8 lm-head: output bf16 (match stock lm-head dtype) instead of fp32 — halves the logits buffer that was inflating vLLM's profiled activation reserve and shrinking the dense KV pool

2026-06-24 15:53:58 +10:00

..

mtp_serve.sh

VALIDATED defaults: gpu-mem 0.82 + seqs 3 (0.88+ over-subscribes/swaps); README corrected with measured KV pool (457k tok, 1.74x) + concurrency curve

2026-06-24 14:30:56 +10:00

patch_fla_shmem.py

qwen3.5-122B-A10B on DGX Spark: vLLM + DFlash + dense-bandwidth stack, one-shot installer

2026-06-24 13:02:35 +10:00

patch_inc_hybrid.py

qwen3.5-122B-A10B on DGX Spark: vLLM + DFlash + dense-bandwidth stack, one-shot installer

2026-06-24 13:02:35 +10:00

patch_int8_lmhead_v3.py

int8 lm-head: output bf16 (match stock lm-head dtype) instead of fp32 — halves the logits buffer that was inflating vLLM's profiled activation reserve and shrinking the dense KV pool

2026-06-24 15:53:58 +10:00

patch_unify2.py

qwen3.5-122B-A10B on DGX Spark: vLLM + DFlash + dense-bandwidth stack, one-shot installer

2026-06-24 13:02:35 +10:00

serve.sh

VALIDATED defaults: gpu-mem 0.82 + seqs 3 (0.88+ over-subscribes/swaps); README corrected with measured KV pool (457k tok, 1.74x) + concurrency curve

2026-06-24 14:30:56 +10:00