[tune] model: ~/.mtplx/models/Youssofal--Qwen3.6-27B-MTPLX-Optimized-Speed
[tune] diagnostic telemetry active; use --no-telemetry for clean speed comparison
[tune] close heavy apps now for cleaner results before measurements start
[tune] fans may get loud during tuning and will be restored afterward
[max] fan tool: thermalforge (~/.mtplx/bin/thermalforge)
[max] baseline fans: min 7829 RPM, max 7845 RPM
[max] running `thermalforge max`...
[max] waiting 1.0s for the daemon to update target RPM...
[max] target accepted: 7826 RPM; waiting for actual fans to ramp...
[max] fans ramped: actual 7810-7822 RPM; target 7826 RPM
[tune] artifacts: <worktree>/outputs/cli/tune/tune-20260702-161546
[tune] running isolated candidates: AR, D3
[tune] AR (1/2) starting; log: <worktree>/outputs/cli/tune/tune-20260702-161546/ar.log
[tune] AR finished in 8.9s: 30.38 tok/s
[tune] AR telemetry: scope=generation | power pkg=22.9W cpu=3.5W ane=0.0W gpu=19.4W | freq P=4.06GHz M=1.02GHz GPU=1.21GHz | temp core_avg=49.7C core_max=59.7C gpu_avg=36.8C | util P=100.0% M=49.7% GPU=98.1% | fans=7836rpm | samples=4 | window=6.6s
[tune] settling 5.0s before D3
[tune] D3 (2/2) starting; log: <worktree>/outputs/cli/tune/tune-20260702-161546/d3.log
[tune] D3 finished in 4.6s: 75.46 tok/s
[tune] D3 telemetry: scope=generation | power pkg=58.8W cpu=4.0W ane=0.0W gpu=54.1W | freq P=4.09GHz M=1.35GHz GPU=1.58GHz | temp core_avg=53.4C core_max=60.8C gpu_avg=37.3C | util P=100.0% M=47.1% GPU=90.5% | fans=7826rpm | samples=2 | window=2.8s
MTPLX Tune
Results written to <worktree>/outputs/cli/tune/tune-20260702-161546

AR     30.4 tok/s   1.00x
D3     75.5 tok/s   2.48x  BEST
Speed shown is decode tok/s; prefill is tracked separately.

Best for this Mac: D3, 2.48x AR
Not saved: save disabled.

AR: telemetry=scope=generation | power pkg=22.9W cpu=3.5W ane=0.0W gpu=19.4W | freq P=4.06GHz M=1.02GHz GPU=1.21GHz | temp core_avg=49.7C core_max=59.7C gpu_avg=36.8C | util P=100.0% M=49.7% GPU=98.1% | fans=7836rpm | samples=4 | window=6.6s
D3: verify_ms=40.07 acceptance=MTP1 50/53 (94.34%) | MTP2 47/53 (88.68%) | MTP3 42/53 (79.25%)
D3: telemetry=scope=generation | power pkg=58.8W cpu=4.0W ane=0.0W gpu=54.1W | freq P=4.09GHz M=1.35GHz GPU=1.58GHz | temp core_avg=53.4C core_max=60.8C gpu_avg=37.3C | util P=100.0% M=47.1% GPU=90.5% | fans=7826rpm | samples=2 | window=2.8s
artifacts: <worktree>/outputs/cli/tune/tune-20260702-161546