This release adds 2 notable features for engineering teams evaluating rollout.
✓ No known CVEs patched in this version
Topics
+7 more
Summary
AI summaryDense weight warm-up reduces first-token latency on large models from ~18 s to ~1 s.
Changes in this release
| Type | Severity | Summary | CVE |
|---|---|---|---|
| Feature | Low |
Adds Dense weight warm-up toggle in Settings and `--no-warm-dense` CLI option to opt out. Adds Dense weight warm-up toggle in Settings and `--no-warm-dense` CLI option to opt out. Source: granite4.1:30b@2026-07-19-audit Confidence: low |
— |
| Performance | Medium |
Dense weight warm-up reduces first token latency from ~18 s to ~1 s for large models. Dense weight warm-up reduces first token latency from ~18 s to ~1 s for large models. Source: llm_adapter@2026-07-19 Confidence: high |
— |
Full changelog
Dense weight warm-up
Models larger than device RAM no longer pay a slow-start: the non-expert weights (embeddings,
attention, norms, lm_head), which stay mmap-resident while experts stream, are now page-cached with
one sequential sweep at load instead of faulting in lazily during the first tokens.
Measured (OnePlus 15R, 11.4 GB RAM): on gpt-oss-120b (~5.2× RAM) the first token drops from
~18 s to ~1 s and the first-five-token wall average improves ~20×. Smaller-dense models (Qwen, Gemma)
are neutral — hit rate and cache budget unchanged, streaming path byte-identical.
On by default; opt out via the Dense weight warm-up toggle in Settings, or --no-warm-dense
on the CLI. Details and per-token data in docs/warmup-analysis.md and
docs/bench-data/2026-07-14-warmup/.
Install
app-dev-debug.apk is the sideload (dev) build for arm64-v8a. It reads models adb-pushed to
/sdcard/Download or /data/local/tmp; installing it does not touch already-downloaded models.
Weekly OSS security release digest.
The CVE patches and breaking changes that affected production tools this week. One email, every Sunday.
No spam, unsubscribe anytime.
Share this release
About BigMoeOnEdge
All releases →Related context
Related tools
Beta — feedback welcome: [email protected]