Jackrong
MLX Community

Qwen3.5-27B (Claude Opus 4.6 Distill) benchmark on an Apple's logo.M5 Pro · 64 GB

<- Runs

Prompt tokens

40,960

Generation tokens

10,240

Trials passed

10/10

Verified

17.2 tok/s

127.3 tok/s

Peak memory

18.00/64 GB

Runs poorly

Trials

Decode / Prefill Speeds

Metadata

metadata.json
{
"runId": "run_0d71947e-1e3e-44c6-8dd1-780e17144c8d",
"bundleId": "mlx-qwen3.5-27b-claude-4.6-opus-distilled-ml-c86058",
"status": "verified",
"promptTokens": 40960,
"completionTokens": 10240,
"contextLength": 5120,
"harness": {
"version": "0.1.21",
"gitSha": "unknown"
},
"runtime": {
"name": "mlx_lm",
"version": "0.31.3",
"buildFlags": null
},
"model": {
"displayName": "Qwen3.5-27B (Claude Opus 4.6 Distill)",
"format": "mlx",
"quant": "4bit",
"architecture": "qwen3_5",
"source": "mlx-community/Qwen3.5-27B-Claude-4.6-Opus-Distilled-MLX-4bit",
"fileSizeBytes": 15133043513,
"lab": {
"name": "Jackrong",
"slug": "jackrong"
},
"quantizedBy": {
"name": "MLX Community",
"slug": "mlx-community"
}
},
"device": {
"cpu": "Apple M5 Pro",
"cpuCores": 18,
"gpu": "Apple M5 Pro",
"gpuCores": 20,
"gpuCount": 1,
"ramGb": 64,
"osName": "macOS",
"osVersion": "26.6"
},
"decodeTpsMean": 17.2,
"prefillTpsMean": 127.3,
"ttftP50Ms": 32234.96,
"idleTpsMean": 3599,
"peakRssMb": 18432,
"trialsPassed": 10,
"trialsTotal": 10,
"runnabilityScore": 0.45608928571428575,
"bundleSha256": "6587a729f4b78527b29f9c592dd13d5a0fed86ea2f614a9bd853fd4b03271874",
"createdAt": "2026-08-04T11:43:19.404Z"
}