[joseedson18jc]

Problem and result

On the M1/M2 family the routed experts of a Q2_0 pack run with the gate/up tile 32,1,4 and the down tile 8,1,4 (the Q2_0 down is moe_down_142). On an M1 Max these are not the fastest tiles that moe.metal already instantiates. expert_parity --bench (Q2_0 gate/up + down, synthetic, M1 Max 64 GB):

tokens	32,1,4 + 8,1,4 (default)	8,1,2 + 4,2,4
1	114 us a layer	94	-17%
2	214	167	-22%
3	318	242	-24%

Live (Qwen3.8-Flash-Next Q2_0, --spec 2 + MTP, through serve/server.py) this is +5-7% greedy decode. This PR adds the tiles as an opt-in knob and does not change any default.

Related discussion

#4238 does the same for the K quants on M1/M2 (gate/up 8,1,2) and leaves Q2_0 on the M5 tile. This is the Q2_0 counterpart, as an env knob rather than a default, because it was measured on one machine.

Approach

STRATA_METAL_MOE_GU / STRATA_METAL_MOE_DN ("L,R,S") in metal/src/engine/window.cpp, read once:

• Both expert windows read it: the decode moe block and mtp_encode.
• gx_gu / gx_dn and the threadgroup widths follow the chosen tile, since they are computed from the same variables.
• Only tiles in GU_SET / DN_SET (the ones moe.metal instantiates) are accepted. Anything else is refused with a log line, and the default stays, so a typo cannot fail pipeline creation.
• The chosen tile is logged once (strata metal: STRATA_METAL_MOE_GU=8,1,2).
• Unset, the code path and the tiles are unchanged.

Validation

Passed:

• expert_parity --gu Q2_0 --dn Q2_0 --ntok {1,2,3} --gucfg 8,1,2 --dncfg 4,2,4: y rel L2 5.3-6.3e-8, PASS (8,1,4 + 4,2,4 also passes).
• expert_parity --bench per config, two runs each, table above. --sweep picks 8,1,2 / 8,1,4 for gate/up and 4,2,4 for down at every token count.
• The parity suite on this change (built on 0.1.51 + the [Engine] metal: the request's sampling on the Mac engine (temperature, top_p, top_k, min_p, seed, penalties, logit_bias) #3301 port, M1 Max, macOS 27): 16/17. fuse_parity fails on router_route, identically without the change (the known M1 Max routing issue).
• Live A/B, bracketed B C B C, the same binary with the knob off/on. Six unseen prompts, 450 tokens each, speed from the server's /metrics, no other traffic:
  • greedy: off 32.7 / 34.0, on 35.8 / 35.7 tok/s.
  • sampled ([Engine] metal: the request's sampling on the Mac engine (temperature, top_p, top_k, min_p, seed, penalties, logit_bias) #3301 sampler): off 31.1 / 32.4, on 33.8 / 33.3 tok/s.
  • The worst "on" arm beats the best "off" arm in both modes.
• This branch (main + the commit) builds strata and expert_parity with no new warnings.

Failed:

• none observed

Not run:

• An engine run of this exact branch: the live A/B binary was 0.1.51 + the [Engine] metal: the request's sampling on the Mac engine (temperature, top_p, top_k, min_p, seed, penalties, logit_bias) #3301 sampler port, and this commit applies to main unchanged.
• Any machine other than the M1 Max, or a pack other than Q2_0 (the K quants are [Engine] Metal: faster routed-expert gate/up tiles on M1/M2 (8,1,2 for the K quants) #4238's domain).
• CUDA / HIP / SYCL: not touched.

Limits

• Measured on one M1 Max 64 GB with one Q2_0 pack. M2, M3 and M4 are unmeasured, so this stays opt-in rather than an M1/M2 default.
• If a maintainer wants it as a default, the natural gate is the same M1/M2 check as [Engine] Metal: faster routed-expert gate/up tiles on M1/M2 (8,1,2 for the K quants) #4238's.