[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 7818 RPM, max 7829 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 7822-7829 RPM; target 7826 RPM
[tune] artifacts: <worktree>/outputs/cli/tune/tune-20260702-161717
[tune] running isolated candidates: AR, D3
[tune] AR (1/2) starting; log: <worktree>/outputs/cli/tune/tune-20260702-161717/ar.log
[tune] AR finished in 8.9s: 30.33 tok/s
[tune] AR telemetry: scope=generation | power pkg=23.8W cpu=3.6W ane=0.0W gpu=20.0W | freq P=4.05GHz M=0.69GHz GPU=1.23GHz | temp core_avg=57.0C core_max=66.2C gpu_avg=38.7C | util P=100.0% M=49.3% GPU=98.0% | fans=7823rpm | samples=4 | window=6.6s
[tune] settling 5.0s before D3
[tune] D3 (2/2) starting; log: <worktree>/outputs/cli/tune/tune-20260702-161717/d3.log
[tune] D3 finished in 4.4s: 81.64 tok/s
[tune] D3 telemetry: scope=generation | power pkg=66.1W cpu=3.6W ane=0.0W gpu=62.8W | freq P=4.03GHz M=0.67GHz GPU=1.60GHz | temp core_avg=59.2C core_max=66.9C gpu_avg=39.2C | util P=100.0% M=50.0% GPU=100.0% | fans=7828rpm | samples=2 | window=2.6s
MTPLX Tune
Results written to <worktree>/outputs/cli/tune/tune-20260702-161717

AR     30.3 tok/s   1.00x
D3     81.6 tok/s   2.69x  BEST
Speed shown is decode tok/s; prefill is tracked separately.

Best for this Mac: D3, 2.69x AR
Not saved: save disabled.

AR: telemetry=scope=generation | power pkg=23.8W cpu=3.6W ane=0.0W gpu=20.0W | freq P=4.05GHz M=0.69GHz GPU=1.23GHz | temp core_avg=57.0C core_max=66.2C gpu_avg=38.7C | util P=100.0% M=49.3% GPU=98.0% | fans=7823rpm | samples=4 | window=6.6s
D3: verify_ms=36.44 acceptance=MTP1 50/53 (94.34%) | MTP2 47/53 (88.68%) | MTP3 42/53 (79.25%)
D3: telemetry=scope=generation | power pkg=66.1W cpu=3.6W ane=0.0W gpu=62.8W | freq P=4.03GHz M=0.67GHz GPU=1.60GHz | temp core_avg=59.2C core_max=66.9C gpu_avg=39.2C | util P=100.0% M=50.0% GPU=100.0% | fans=7828rpm | samples=2 | window=2.6s
artifacts: <worktree>/outputs/cli/tune/tune-20260702-161717