[timnevits]

Adds Gen4 x16 benchmark results for one Arc Pro B65 on official v0.1.41 (fb58e0db), with no local engine patches. Original full 512-expert Flash-Next IQ2_XS, INT8 KV, protected pinned expert mirror, GPU-only execution and confidence-limited MTP4. Uses the documented STRATA_SYCL_SPIN_MAX=20000 build option to preserve the xe wait bound on this host with separate Intel display and compute GPUs.

At 7K input/640 output, 8K/512 measures 43.52 decode tok/s and 262K/512 measures 42.39. At 8K, prefill 4096 changes median 7K TTFT from 18.20 to 7.62s. Normal serving keeps prefill 512; the report details per-workload decode differences and changed output hashes.

Three fresh runs per profile in each of two suites: matched 512/7000-input, 640-output tasks and separate public 20/2185-input, 256-output fixtures. Includes repetitions, native prompt/decode timings, medians/ranges, startup/cache state, exact hardware/build/artifact/configuration pins and portable reproduction scripts. Full-window capacity checks are separate.

Native reference, API, lifecycle and hardware checks pass for the measured profiles. The full registered kernel suite has four classified failures (26/30 pass); the report explains those limits and separate passing IQ/model-stream checks. No engine patches or tolerance changes.

Codex, an OpenAI AI agent, ran the tests and drafted and submitted this report with my approval. The measurements come from my B65 hardware.

[Niko1221]

Thanks for the pull request.

We're resetting the tracker for 0.2.0-beta. There are more than 1,300 open issues and pull requests, too many for us to review properly, so we're closing everything that isn't being worked on right now. Closed here does not mean rejected.

To reopen it takes about two minutes:

1. Rebase on main if it no longer applies, and re-run your checks on 0.2.0-beta.
2. Start the title with its area: [Engine], [GUI], [TUI], [Benchmark], [Docs] or [Other].
3. Fill in the pull request template: the problem and result, and the validation (what you ran, on which revision, what you saw).
4. Comment /reopen. A check runs and either reopens the pull request or lists what is still missing.

The rules are in CONTRIBUTING.md.