vrr/BigMoeOnEdge
2
0
Fork 0
mirror of https://github.com/Helldez/BigMoeOnEdge.git synced 2026-10-19 15:33:46 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 2
BigMoeOnEdge/scripts
History
Exact
Raffaele e232c96447
docs(readme): rewrite the README around the adaptive streaming optimizer (#296) 
* docs(readme): rewrite the README around the adaptive streaming optimizer

The README opens with the promo video and says what the engine is: an adaptive
streaming optimizer for big local AI agents on any hardware, OS, model and
quantization, behind a local OpenAI-compatible server. Ten measured runs are
shown as examples, not limits, and three new diagrams explain how one token
runs, how the planner adapts and where the memory goes. Settings, architecture
and benchmark tables move out to the docs that already hold them; the three
GIFs and the older demo video go.

* docs(readme): the Qwen3.6-35B laptop row spans the measured 18 to 24.2 tok/s

* docs(readme): link the promo video
2026-10-19 08:36:15 +02:00
..
android-gpu-deps.sh feat(build): compute backends as modules beside the engine, on every OS (#263) 2026-10-18 12:15:28 +02:00
bench-analyze.py chore(scripts): consolidate the bench drivers, prune the retired framing 2026-07-17 10:41:13 +02:00
bench-lib.ps1 chore(scripts): consolidate the bench drivers, prune the retired framing 2026-07-17 10:41:13 +02:00
bench-matrix-rework.ps1 chore(scripts): consolidate the bench drivers, prune the retired framing 2026-07-17 10:41:13 +02:00
bench-matrix.ps1 chore(scripts): consolidate the bench drivers, prune the retired framing 2026-07-17 10:41:13 +02:00
bench-pr23-c2000.ps1 chore(scripts): consolidate the bench drivers, prune the retired framing 2026-07-17 10:41:13 +02:00
bench-pr23-summary.py chore(scripts): consolidate the bench drivers, prune the retired framing 2026-07-17 10:41:13 +02:00
bench-prefetch.ps1 chore(scripts): consolidate the bench drivers, prune the retired framing 2026-07-17 10:41:13 +02:00
bench-report.sh feat(plan): the hardware planner, the parameter table and the macOS port (#214) 2026-10-13 19:37:10 +02:00
bench-run.sh Mark the two doc-invoked host scripts executable (#164) 2026-08-25 11:41:24 +02:00
bench-warmonly.sh Mark the two doc-invoked host scripts executable (#164) 2026-08-25 11:41:24 +02:00
build-android.ps1 feat(build): compute backends as modules beside the engine, on every OS (#263) 2026-10-18 12:15:28 +02:00
build-hexagon-android.sh feat(build): compute backends as modules beside the engine, on every OS (#263) 2026-10-18 12:15:28 +02:00
build-host.sh feat(plan): the hardware planner, the parameter table and the macOS port (#214) 2026-10-13 19:37:10 +02:00
build-ios.sh feat(build): compute backends as modules beside the engine, on every OS (#263) 2026-10-18 12:15:28 +02:00
check-version.sh fix(build): the engine reports the version it was released as, and CI checks it (#224) 2026-10-13 19:29:35 +02:00
decode-analyze.py feat(trace): add layer-granularity compute trace (--compute-trace-layers) 2026-07-19 09:21:39 +02:00
gptoss-matrix.sh Mark the two doc-invoked host scripts executable (#164) 2026-08-25 11:41:24 +02:00
gptoss-mmap.sh Mark the two doc-invoked host scripts executable (#164) 2026-08-25 11:41:24 +02:00
humaneval-bench.py feat(plan): the hardware planner, the parameter table and the macOS port (#214) 2026-10-13 19:37:10 +02:00
inspect-hf-moe-release.py chore: retire the files and the words the architecture table left behind (#277) 2026-10-18 11:33:12 +02:00
make-tiny-moe.py feat: --decide, choose from a list from one prefill, with no decode (#201) 2026-09-28 12:01:25 +02:00
readme-diagrams.py docs(readme): rewrite the README around the adaptive streaming optimizer (#296) 2026-10-19 08:36:15 +02:00
rented-box.sh feat(build): compute backends as modules beside the engine, on every OS (#263) 2026-10-18 12:15:28 +02:00
route-analyze.py feat(moe): --drop-cold-experts — spend quality only where it buys I/O (#95) 2026-07-22 17:21:55 +02:00
route-drop-replay.py feat(moe): --drop-cold-experts — spend quality only where it buys I/O (#95) 2026-07-22 17:21:55 +02:00
route-replay.py fix(tools): make the diagnostics fail instead of reporting a confident wrong number 2026-07-20 12:25:58 +02:00
route-viewer.py refactor: no hardcoding, no regex, the duplication removed and the server and app hardened (#262) 2026-10-18 10:49:39 +02:00
stage-hexagon-jnilibs.ps1 feat(build): compute backends as modules beside the engine, on every OS (#263) 2026-10-18 12:15:28 +02:00
tinymmlu-bench.py feat(plan): the hardware planner, the parameter table and the macOS port (#214) 2026-10-13 19:37:10 +02:00
trace_io.py chore(scripts): consolidate the bench drivers, prune the retired framing 2026-07-17 10:41:13 +02:00