Ettore Di Giacinto 1431f72b92 docs(paged): regenerate decode plots (3-way) from re-measured data + overview ...

Rebuild the two committed decode plots from the re-measured CSV and add a combined
overview. Three series per the comparison that matters: llama.cpp (standard) vs
vLLM vs LocalAI's llama.cpp patches; x-over-standard called out at npl128. bf16-tau
stays out of the plot (it remains in the CSV + the README table as the opt-in row).

Assisted-by: Claude:opus-4.8 [Claude Code]
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>

2026-06-27 22:20:12 +00:00

..

final_benchmark.csv

docs(paged): re-measure DGX benchmarks on one harness (stock/patched/bf16-tau)

2026-06-27 22:05:59 +00:00

LOCALAI_LLAMACPP_BACKEND_PLAN.md

chore(paged): keep patches/ patch-only; README to backend root, docs to docs/

2026-06-27 13:20:05 +00:00

PAGED_BITEXACT_NOTE.md

chore(paged): keep patches/ patch-only; README to backend root, docs to docs/

2026-06-27 13:20:05 +00:00

paged-burst-bench.cpp

chore(paged): keep patches/ patch-only; README to backend root, docs to docs/

2026-06-27 13:20:05 +00:00

paged-reclaim-unit.cpp

chore(paged): keep patches/ patch-only; README to backend root, docs to docs/

2026-06-27 13:20:05 +00:00

PATCH_MAINTENANCE.md

chore(paged): keep patches/ patch-only; README to backend root, docs to docs/

2026-06-27 13:20:05 +00:00

qwen36_decode_overview.png

docs(paged): regenerate decode plots (3-way) from re-measured data + overview

2026-06-27 22:20:12 +00:00

qwen36_dense_decode_vs_npl.png

docs(paged): regenerate decode plots (3-way) from re-measured data + overview

2026-06-27 22:20:12 +00:00

qwen36_moe_decode_vs_npl.png

docs(paged): regenerate decode plots (3-way) from re-measured data + overview

2026-06-27 22:20:12 +00:00

UPSTREAM_LAYER2_SCOPE.md

chore(paged): keep patches/ patch-only; README to backend root, docs to docs/

2026-06-27 13:20:05 +00:00