Jackrong
Jackrong

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

<- Runs

Prompt tokens

40,960

Generation tokens

10,240

Trials passed

10/10

Verified

82.6 tok/s

937.6 tok/s

Peak memory

21.00/64 GB

Runs well

Trials

Decode / Prefill Speeds

Metadata

metadata.json
{
"runId": "run_3d4484f6-4043-4ac6-8642-b5df006ecd20",
"bundleId": "mlx-mlx-qwen3.5-35b-a3b-claude-4.6-opus-reas-151be7",
"status": "verified",
"promptTokens": 40960,
"completionTokens": 10240,
"contextLength": 5120,
"harness": {
"version": "0.1.20",
"gitSha": "unknown"
},
"runtime": {
"name": "mlx_lm",
"version": "0.31.2",
"buildFlags": null
},
"model": {
"displayName": "Qwen3.5-35B-A3B (Claude Opus 4.6 Distill)",
"format": "mlx",
"quant": "4bit",
"architecture": "qwen3_5_moe",
"source": "Jackrong/MLX-Qwen3.5-35B-A3B-Claude-4.6-Opus-Reasoning-Distilled-4bit",
"fileSizeBytes": 19509023921,
"lab": {
"name": "Jackrong",
"slug": "jackrong"
},
"quantizedBy": {
"name": "Jackrong",
"slug": "jackrong"
}
},
"device": {
"cpu": "Apple M4 Pro",
"cpuCores": 14,
"gpu": "Apple M4 Pro",
"gpuCores": 20,
"gpuCount": 1,
"ramGb": 64,
"osName": "macOS",
"osVersion": "26.3"
},
"decodeTpsMean": 82.6,
"prefillTpsMean": 937.6,
"ttftP50Ms": 4319.85,
"idleTpsMean": 19456,
"peakRssMb": 21504,
"trialsPassed": 10,
"trialsTotal": 10,
"runnabilityScore": 0.7270350000000001,
"bundleSha256": "64ef478ab52cb8aea110434596aa0c667ec4eb534bab38520a7786d55136fa07",
"createdAt": "2026-04-14T20:05:04.517Z"
}