Jackrong
Jackrong

Qwen3.5-35B-A3B (Claude Opus 4.6 Distill) benchmark on an Apple's logo.M4 Max · 64 GB

<- Runs

Prompt tokens

12,288

Generation tokens

3,072

Trials passed

3/3

Verified

84.4 tok/s

1,483.8 tok/s

Peak memory

37.00/64 GB

Runs ok

Trials

Decode / Prefill Speeds

Metadata

metadata.json
{
"runId": "run_66046fdc-fb2d-4678-b8d6-fffac5f4eecd",
"bundleId": "mlx-mlx-qwen3.5-35b-a3b-claude-4.6-opus-reas-e2f9b8",
"status": "verified",
"promptTokens": 12288,
"completionTokens": 3072,
"contextLength": 5120,
"harness": {
"version": "0.1.16",
"gitSha": "unknown"
},
"runtime": {
"name": "mlx_lm",
"version": "0.31.1",
"buildFlags": null
},
"model": {
"displayName": "Qwen3.5-35B-A3B (Claude Opus 4.6 Distill)",
"format": "mlx",
"quant": "8bit",
"architecture": "qwen3_5_moe",
"source": "Jackrong/MLX-Qwen3.5-35B-A3B-Claude-4.6-Opus-Reasoning-Distilled-8bit",
"fileSizeBytes": 36828222525,
"lab": {
"name": "Jackrong",
"slug": "jackrong"
},
"quantizedBy": {
"name": "Jackrong",
"slug": "jackrong"
}
},
"device": {
"cpu": "Apple M4 Max",
"cpuCores": 16,
"gpu": "Apple M4 Max",
"gpuCores": 40,
"gpuCount": 1,
"ramGb": 64,
"osName": "macOS",
"osVersion": "26.3.1"
},
"decodeTpsMean": 84.4,
"prefillTpsMean": 1483.8,
"ttftP50Ms": 2791.31,
"idleTpsMean": 35840,
"peakRssMb": 37888,
"trialsPassed": 3,
"trialsTotal": 3,
"runnabilityScore": 0.6530221428571429,
"bundleSha256": "eb3b26b17e24be03948530ff2364ebfcdfa893b1e831a4707a098b80ba4fbb96",
"createdAt": "2026-03-31T01:36:12.794Z"
}