1{"backend": "titan", "capacity": 16, "policy": "lru", "predictor": "none", "m": 8, "prefetch_workers": 2, "nocache": true, "max_tokens": 64, "split": "heldout", "tag": "cold_nvme", "model": "/Users/simon-pierreboucher/.cache/huggingface/hub/models--mlx-community--OLMoE-1B-7B-0125-Instruct-4bit/snapshots/e23844197887b031e7ddddbb0b8959c5a6853a7b", "cmd": "benchmarks/scripts/bench_decode.py --model /Users/simon-pierreboucher/.cache/huggingface/hub/models--mlx-community--OLMoE-1B-7B-0125-Instruct-4bit/snapshots/e23844197887b031e7ddddbb0b8959c5a6853a7b --backend titan --capacity 16 --predictor none --m 8 --nocache --traces benchmarks/raw/gateC/traces_olmoe_v2_laptop --families general,code,reasoning --split heldout --repeats 2 --max-tokens 64 --tag cold_nvme --out benchmarks/raw/gateC/cold_nvme_laptop/sweep.jsonl", "env": {"host": "MacBookPro-de-simon-pierre-2.local", "chip": "Apple M5 Max", "ram_gb": 48, "macos": "27.0", "python": "3.12.13", "mlx": "0.32.4", "mlx_lm": "0.32.2", "numpy": "2.5.4", "git": "258ba81"}, "seed": 0, "warmup": true, "repeat": -1, "family": "reasoning", "prompt": "Q: A tank fills in 6 hours with pipe A and 9 hours with pipe", "ts": "2026-11-01T01:47:04", "prompt_tokens": 32, "gen_tokens": 64, "ttft_ms": 3041.2979170214385, "prompt_tps": 10.538982502473754, "gen_tps": 9.615565496623349, "decode_tps_wall": 9.465435344104892, "itl_ms_p50": 102.56666608620435, "itl_ms_p95": 147.70795905496925, "itl_ms_p99": 167.03716607298702, "itl_ms_mean": 105.64754431742051, "total_s": 9.69709320901893, "peak_mem_gb": 2.090173904, "text": "If pipe A fills the tank in 6 hours, then the rate of pipe A is 1 / 6 of the tank per hour. If pipe B fills the tank in 9 hours, then the rate of pipe B is 1 / 9 of the tank per hour. The combined rat", "cache_hit_rate": 0.4844649021864212, "cache_misses": 5824, "cache_hits": 5473, "stall_s": 8.191949911764823, "hits_device": 0, "layers_complete": 0, "prefetch": null, "ssd_bytes": 20610809856, "ssd_read_s": 7.725407627178356}2{"backend": "titan", "capacity": 16, "policy": "lru", "predictor": "none", "m": 8, "prefetch_workers": 2, "nocache": true, "max_tokens": 64, "split": "heldout", "tag": "cold_nvme", "model": "/Users/simon-pierreboucher/.cache/huggingface/hub/models--mlx-community--OLMoE-1B-7B-0125-Instruct-4bit/snapshots/e23844197887b031e7ddddbb0b8959c5a6853a7b", "cmd": "benchmarks/scripts/bench_decode.py --model /Users/simon-pierreboucher/.cache/huggingface/hub/models--mlx-community--OLMoE-1B-7B-0125-Instruct-4bit/snapshots/e23844197887b031e7ddddbb0b8959c5a6853a7b --backend titan --capacity 16 --predictor none --m 8 --nocache --traces benchmarks/raw/gateC/traces_olmoe_v2_laptop --families general,code,reasoning --split heldout --repeats 2 --max-tokens 64 --tag cold_nvme --out benchmarks/raw/gateC/cold_nvme_laptop/sweep.jsonl", "env": {"host": "MacBookPro-de-simon-pierre-2.local", "chip": "Apple M5 Max", "ram_gb": 48, "macos": "27.0", "python": "3.12.13", "mlx": "0.32.4", "mlx_lm": "0.32.2", "numpy": "2.5.4", "git": "258ba81"}, "seed": 0, "warmup": true, "repeat": -1, "family": "code", "prompt": "Write a C function that computes the dot product of two floa", "ts": "2026-11-01T01:47:11", "prompt_tokens": 24, "gen_tokens": 64, "ttft_ms": 1772.822791012004, "prompt_tps": 13.540661227523396, "gen_tps": 14.592826924160933, "decode_tps_wall": 14.364850714221733, "itl_ms_p50": 64.1560829244554, "itl_ms_p95": 111.17766599636525, "itl_ms_p99": 147.35658396966755, "itl_ms_mean": 69.61436772955552, "total_s": 6.158527957974002, "peak_mem_gb": 1.930492816, "text": "#include <arm_neon.h>\n\nfloat dot_product_neon(const float *a, const float *b, int len) {\n // TODO: Implement the dot product computation using NEON intrinsics\n // Start by loading the arrays int", "cache_hit_rate": 0.6523887973640856, "cache_misses": 3587, "cache_hits": 6732, "stall_s": 4.940353738376871, "hits_device": 0, "layers_complete": 0, "prefetch": null, "ssd_bytes": 12694192128, "ssd_read_s": 4.6701676497468725}3{"backend": "titan", "capacity": 16, "policy": "lru", "predictor": "none", "m": 8, "prefetch_workers": 2, "nocache": true, "max_tokens": 64, "split": "heldout", "tag": "cold_nvme", "model": "/Users/simon-pierreboucher/.cache/huggingface/hub/models--mlx-community--OLMoE-1B-7B-0125-Instruct-4bit/snapshots/e23844197887b031e7ddddbb0b8959c5a6853a7b", "cmd": "benchmarks/scripts/bench_decode.py --model /Users/simon-pierreboucher/.cache/huggingface/hub/models--mlx-community--OLMoE-1B-7B-0125-Instruct-4bit/snapshots/e23844197887b031e7ddddbb0b8959c5a6853a7b --backend titan --capacity 16 --predictor none --m 8 --nocache --traces benchmarks/raw/gateC/traces_olmoe_v2_laptop --families general,code,reasoning --split heldout --repeats 2 --max-tokens 64 --tag cold_nvme --out benchmarks/raw/gateC/cold_nvme_laptop/sweep.jsonl", "env": {"host": "MacBookPro-de-simon-pierre-2.local", "chip": "Apple M5 Max", "ram_gb": 48, "macos": "27.0", "python": "3.12.13", "mlx": "0.32.4", "mlx_lm": "0.32.2", "numpy": "2.5.4", "git": "258ba81"}, "seed": 0, "warmup": true, "repeat": -1, "family": "general", "prompt": "Summarize the history of the Rideau Canal in one paragraph.", "ts": "2026-11-01T01:47:19", "prompt_tokens": 14, "gen_tokens": 64, "ttft_ms": 1448.2218329794705, "prompt_tps": 9.668972925355584, "gen_tps": 9.486046026204507, "decode_tps_wall": 9.337841435614223, "itl_ms_p50": 106.86812503263354, "itl_ms_p95": 156.7806249950081, "itl_ms_p99": 170.41387502104044, "itl_ms_mean": 107.09113095303081, "total_s": 8.194963083020411, "peak_mem_gb": 1.742661784, "text": "\n\nThe Rideau Canal, officially opened in 1832, was constructed by the British to connect Ottawa to Montreal via the Ottawa River. It was designed by Colonel John By and took approximately three years ", "cache_hit_rate": 0.48355739400206826, "cache_misses": 4994, "cache_hits": 4676, "stall_s": 6.844891174114309, "hits_device": 0, "layers_complete": 0, "prefetch": null, "ssd_bytes": 17673486336, "ssd_read_s": 6.502651879563928}4