nightshift: [1792082901] start v0.90.0 · Apple M5 (m5+) · 16 GB · macOS 26.4 · 4P+6E cores
nightshift: [1792082901] serving /Users/samuelcummings/nightshift-m1m4-build/web-bench on http://127.0.0.1:47321/
nightshift: [1792082901] webkit: BackgroundWebContentRunningBoardThrottlingEnabled=false, HiddenPageDOMTimerThrottlingAutoIncreases=false, HiddenPageDOMTimerThrottlingEnabled=false, PageVisibilityBasedProcessSuppressionEnabled=false, WebGPUEnabled=true
nightshift: [1792082901] webkit: window occlusion detection off (page stays active behind other windows): true
nightshift: [1792082901] cockpit loading (injected scripts removed: true)
nightshift: [1792082901] quit gate installed (applicationShouldTerminate:)
nightshift: [1792082904] page: {"bench":{"label":"97-loadmem","phases":["env","models"],"seq":0,"t":2921,"type":"start"}}
nightshift: [1792082904] page: {"bench":{"phase":"env","seq":1,"t":2922,"type":"phase_start"}}
nightshift: [1792082904] page: {"bench":{"boot":{"bench_script_ms":225,"dom_content_loaded_ms":358,"engine_ready_ms":256,"load_event_ms":417,"resources":211,"scripts":175,"shell_mounted_ms":425,"transfer_kb":27},"cached_models":["Llama-3.2-1B-Instruct-q4f16_1-MLC","SmolLM2-1.7B-Instruct-q4f16_1-MLC","Qwen2.5-1.5B-Instruct-q4f16_1-MLC","gemma-2-2b-it-q4f16_1-MLC","Llama-3.2-3B-Instruct-q4f16_1-MLC","Qwen2.5-3B-Instruct-q4f16_1-MLC","Qwen2.5-Coder-3B-Instruct-q4f16_1-MLC","Phi-3.5-mini-instruct-q4f16_1-MLC"],"device_memory":null,"engine":{"cloud_models":50,"device_models":8,"job_kinds":["deliberate","best_of_n","batch","bulk","app_stream"],"lanes":[],"models_in_registry":66},"errors":[],"gpu":{"adapter":true,"api":true,"architecture":"apple","description":"apple","device":"apple","features":["bgra8unorm-storage","core-features-and-limits","depth-clip-control","depth32float-stencil8","float16-renderable","float32-blendable","float32-filterable","float32-renderable","indirect-first-instance","rg11b10ufloat-renderable","shader-f16","texture-compression-astc","texture-compression-astc-sliced-3d","texture-compression-bc","texture-compression-bc-sliced-3d","texture-compression-etc2","texture-formats-tier1","timestamp-query"],"limits":{"maxBufferSize_mb":2048,"maxComputeInvocationsPerWorkgroup":1024,"maxComputeWorkgroupSizeX":1024,"maxComputeWorkgroupStorageSize":32768,"maxStorageBufferBindingSize_mb":2048,"maxStorageBuffersPerShaderStage":44},"vendor":"apple"},"hardware_concurrency":8,"memory":{"free_pct":65,"pressure":"normal","swap_used_bytes":2703360000},"seq":2,"storage":{"persisted":false,"quota_gb":41.2,"usage_mb":11528},"system":{"arch":"aarch64","chip":"Apple M5","chip_family":"m5+","eff_cores":6,"macos":"26.4","mem_bytes":17179869184,"mem_gb":16,"model":"Mac17,4","perf_cores":4},"t":2987,"tier":{"budgetMb":6000,"label":"16 GB","maxGb":16,"maxModels":2,"panelMb":3000},"type":"env","user_agent":"Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/605.1.15 (KHTML, like Gecko)","webllm":{"catalog":["Llama-3.2-1B-Instruct-q4f16_1-MLC","Llama-3.2-3B-Instruct-q4f16_1-MLC","Qwen2.5-1.5B-Instruct-q4f16_1-MLC","Qwen2.5-3B-Instruct-q4f16_1-MLC","Qwen2.5-Coder-3B-Instruct-q4f16_1-MLC","gemma-2-2b-it-q4f16_1-MLC","Phi-3.5-mini-instruct-q4f16_1-MLC","SmolLM2-1.7B-Instruct-q4f16_1-MLC"],"prebuilt":163,"webgpu":true,"worker_ready_s":0.058}}}
nightshift: [1792082904] page: {"bench":{"phase":"env","seconds":0.066,"seq":3,"t":2988,"type":"phase_end"}}
nightshift: [1792082904] page: {"bench":{"phase":"models","seq":4,"t":2990,"type":"phase_start"}}
nightshift: [1792082910] page: {"boot":{"budget_mb":6000,"errors":[],"gpu":{"adapter":true,"api":true,"f16":true,"max_buffer_mb":2048},"jobs_configured":true,"max_models":2,"models":66,"shell_mounted":true,"storage":{"persisted":false,"quota_gb":41.2,"usage_mb":11528},"tauri_globals":false,"tier":"16 GB","webllm":{"prebuilt":163,"worker_webgpu":true}}}
nightshift: [1792082920] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":5,"t":18390,"text":"Start to fetch params","type":"progress"}}
nightshift: [1792082921] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":6,"t":19574,"text":"Loading model from cache[1/42]: 282MB loaded. 20% completed, 15 secs elapsed.","type":"progress"}}
nightshift: [1792082921] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":7,"t":20063,"text":"Loading model from cache[6/42]: 432MB loaded. 30% completed, 16 secs elapsed.","type":"progress"}}
nightshift: [1792082922] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":8,"t":20428,"text":"Loading model from cache[11/42]: 568MB loaded. 40% completed, 16 secs elapsed.","type":"progress"}}
nightshift: [1792082922] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":9,"t":20778,"text":"Loading model from cache[16/42]: 703MB loaded. 50% completed, 17 secs elapsed.","type":"progress"}}
nightshift: [1792082922] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":10,"t":21089,"text":"Loading model from cache[21/42]: 850MB loaded. 60% completed, 17 secs elapsed.","type":"progress"}}
nightshift: [1792082923] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":11,"t":21436,"text":"Loading model from cache[26/42]: 985MB loaded. 70% completed, 17 secs elapsed.","type":"progress"}}
nightshift: [1792082923] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":12,"t":21825,"text":"Loading model from cache[31/42]: 1131MB loaded. 80% completed, 18 secs elapsed.","type":"progress"}}
nightshift: [1792082924] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":13,"t":22175,"text":"Loading model from cache[36/42]: 1267MB loaded. 90% completed, 18 secs elapsed.","type":"progress"}}
nightshift: [1792082924] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":14,"t":22522,"text":"Loading model from cache[42/42]: 1403MB loaded. 100% completed, 18 secs elapsed.","type":"progress"}}
nightshift: [1792082937] page: {"bench":{"decode":[{"completion_tokens":15,"decode_tps":29.9,"e2e_s":0.845,"prefill_tps":123.9,"prompt_tokens":42,"text":"## The Pedal Revolution: A Century of Cycling\n\nThe bicycle, a seemingly simple","ttft_s":0.339,"wall_s":0.852}],"first_call":{"ttft_s":0.469,"wall_s":0.684},"load_s":19.047,"mem_after_load":{"free_pct":43,"pressure":"normal","swap_used_bytes":2694971392},"mem_after_speed":{"free_pct":33,"pressure":"warn","swap_used_bytes":2686582784},"mem_before":{"free_pct":65,"pressure":"normal","swap_used_bytes":2694971392},"model":"gemma-2-2b-it-q4f16_1-MLC","prefill":[{"completion_tokens":23,"decode_tps":20.1,"e2e_s":11.488,"prefill_tps":144.1,"prompt_tokens":1490,"text":"The text describes the architecture and workflow of the Nightshift runtime for inference jobs, emphasizing its use of device and cloud lanes","ttft_s":10.341,"wall_s":11.492}],"seq":15,"t":35575,"type":"model_speed","was_cached":true}}
nightshift: [1792082937] page: {"bench":{"items":0,"model":"gemma-2-2b-it-q4f16_1-MLC","seconds":0,"seq":16,"t":35579,"type":"suite_done"}}
nightshift: [1792082954] page: {"bench":{"label":"load Llama-3.2-3B-Instruct-q4f16_1-MLC","seq":17,"t":52321,"text":"Start to fetch params","type":"progress"}}
nightshift: [1792082954] page: {"bench":{"label":"load Llama-3.2-3B-Instruct-q4f16_1-MLC","seq":18,"t":52940,"text":"Loading model from cache[1/58]: 188MB loaded. 10% completed, 17 secs elapsed.","type":"progress"}}
nightshift: [1792082955] page: {"bench":{"label":"load Llama-3.2-3B-Instruct-q4f16_1-MLC","seq":19,"t":53360,"text":"Loading model from cache[8/58]: 374MB loaded. 21% completed, 17 secs elapsed.","type":"progress"}}
nightshift: [1792082955] page: {"bench":{"label":"load Llama-3.2-3B-Instruct-q4f16_1-MLC","seq":20,"t":53704,"text":"Loading model from cache[14/58]: 536MB loaded. 31% completed, 17 secs elapsed.","type":"progress"}}
nightshift: [1792082955] page: {"bench":{"label":"load Llama-3.2-3B-Instruct-q4f16_1-MLC","seq":21,"t":54061,"text":"Loading model from cache[20/58]: 698MB loaded. 40% completed, 18 secs elapsed.","type":"progress"}}
nightshift: [1792082956] page: {"bench":{"label":"load Llama-3.2-3B-Instruct-q4f16_1-MLC","seq":22,"t":54513,"text":"Loading model from cache[27/58]: 884MB loaded. 51% completed, 18 secs elapsed.","type":"progress"}}
nightshift: [1792082956] page: {"bench":{"label":"load Llama-3.2-3B-Instruct-q4f16_1-MLC","seq":23,"t":54909,"text":"Loading model from cache[33/58]: 1046MB loaded. 60% completed, 19 secs elapsed.","type":"progress"}}
nightshift: [1792082957] page: {"bench":{"label":"load Llama-3.2-3B-Instruct-q4f16_1-MLC","seq":24,"t":55347,"text":"Loading model from cache[39/58]: 1208MB loaded. 70% completed, 19 secs elapsed.","type":"progress"}}
nightshift: [1792082957] page: {"bench":{"label":"load Llama-3.2-3B-Instruct-q4f16_1-MLC","seq":25,"t":55797,"text":"Loading model from cache[46/58]: 1400MB loaded. 81% completed, 20 secs elapsed.","type":"progress"}}
nightshift: [1792082958] page: {"bench":{"label":"load Llama-3.2-3B-Instruct-q4f16_1-MLC","seq":26,"t":56225,"text":"Loading model from cache[52/58]: 1562MB loaded. 90% completed, 20 secs elapsed.","type":"progress"}}
nightshift: [1792082958] page: {"bench":{"label":"load Llama-3.2-3B-Instruct-q4f16_1-MLC","seq":27,"t":56674,"text":"Loading model from cache[58/58]: 1724MB loaded. 100% completed, 20 secs elapsed.","type":"progress"}}
nightshift: [1792082974] page: {"bench":{"decode":[{"completion_tokens":15,"decode_tps":24.4,"e2e_s":1.105,"prefill_tps":111.1,"prompt_tokens":54,"text":"The bicycle, a ubiquitous mode of transportation and recreation, has a rich and fascinating","ttft_s":0.486,"wall_s":1.105}],"first_call":{"ttft_s":0.505,"wall_s":0.754},"load_s":20.586,"mem_after_load":{"free_pct":40,"pressure":"normal","swap_used_bytes":2661416960},"mem_after_speed":{"free_pct":34,"pressure":"normal","swap_used_bytes":2627862528},"mem_before":{"free_pct":33,"pressure":"warn","swap_used_bytes":2686582784},"model":"Llama-3.2-3B-Instruct-q4f16_1-MLC","prefill":[{"completion_tokens":23,"decode_tps":18.6,"e2e_s":14.429,"prefill_tps":110.6,"prompt_tokens":1458,"text":"The text is a repetitive and redundant list of sections about the Nightshift runtime scheduling inference jobs across device and cloud lanes.","ttft_s":13.188,"wall_s":14.431}],"seq":28,"t":73016,"type":"model_speed","was_cached":true}}
nightshift: [1792082974] page: {"bench":{"items":0,"model":"Llama-3.2-3B-Instruct-q4f16_1-MLC","seconds":0,"seq":29,"t":73023,"type":"suite_done"}}
nightshift: [1792082990] page: {"bench":{"label":"load Phi-3.5-mini-instruct-q4f16_1-MLC","seq":30,"t":88802,"text":"Start to fetch params","type":"progress"}}
nightshift: [1792082991] page: {"bench":{"label":"load Phi-3.5-mini-instruct-q4f16_1-MLC","seq":31,"t":89457,"text":"Loading model from cache[8/83]: 215MB loaded. 10% completed, 16 secs elapsed.","type":"progress"}}
nightshift: [1792082991] page: {"bench":{"label":"load Phi-3.5-mini-instruct-q4f16_1-MLC","seq":32,"t":90020,"text":"Loading model from cache[16/83]: 413MB loaded. 20% completed, 17 secs elapsed.","type":"progress"}}
nightshift: [1792082992] page: {"bench":{"label":"load Phi-3.5-mini-instruct-q4f16_1-MLC","seq":33,"t":90739,"text":"Loading model from cache[25/83]: 636MB loaded. 30% completed, 17 secs elapsed.","type":"progress"}}
nightshift: [1792082993] page: {"bench":{"label":"load Phi-3.5-mini-instruct-q4f16_1-MLC","seq":34,"t":91527,"text":"Loading model from cache[33/83]: 839MB loaded. 40% completed, 18 secs elapsed.","type":"progress"}}
nightshift: [1792082994] page: {"bench":{"label":"load Phi-3.5-mini-instruct-q4f16_1-MLC","seq":35,"t":92205,"text":"Loading model from cache[41/83]: 1037MB loaded. 50% completed, 19 secs elapsed.","type":"progress"}}
nightshift: [1792082994] page: {"bench":{"label":"load Phi-3.5-mini-instruct-q4f16_1-MLC","seq":36,"t":92881,"text":"Loading model from cache[49/83]: 1235MB loaded. 60% completed, 20 secs elapsed.","type":"progress"}}
nightshift: [1792082995] page: {"bench":{"label":"load Phi-3.5-mini-instruct-q4f16_1-MLC","seq":37,"t":93595,"text":"Loading model from cache[58/83]: 1447MB loaded. 70% completed, 20 secs elapsed.","type":"progress"}}
nightshift: [1792082996] page: {"bench":{"label":"load Phi-3.5-mini-instruct-q4f16_1-MLC","seq":38,"t":94222,"text":"Loading model from cache[66/83]: 1649MB loaded. 80% completed, 21 secs elapsed.","type":"progress"}}
nightshift: [1792082996] page: {"bench":{"label":"load Phi-3.5-mini-instruct-q4f16_1-MLC","seq":39,"t":94791,"text":"Loading model from cache[74/83]: 1848MB loaded. 90% completed, 21 secs elapsed.","type":"progress"}}
nightshift: [1792082997] page: {"bench":{"label":"load Phi-3.5-mini-instruct-q4f16_1-MLC","seq":40,"t":95538,"text":"Loading model from cache[83/83]: 2051MB loaded. 100% completed, 22 secs elapsed.","type":"progress"}}
nightshift: [1792083018] page: {"bench":{"decode":[{"completion_tokens":15,"decode_tps":19.2,"e2e_s":1.476,"prefill_tps":89.9,"prompt_tokens":62,"text":"The bicycle, a marvel of human ingenuity, has under","ttft_s":0.69,"wall_s":1.478}],"first_call":{"ttft_s":0.68,"wall_s":1.018},"load_s":21.983,"mem_after_load":{"free_pct":38,"pressure":"warn","swap_used_bytes":2619473920},"mem_after_speed":{"free_pct":28,"pressure":"warn","swap_used_bytes":2611085312},"mem_before":{"free_pct":34,"pressure":"normal","swap_used_bytes":2627862528},"model":"Phi-3.5-mini-instruct-q4f16_1-MLC","prefill":[{"completion_tokens":23,"decode_tps":18.3,"e2e_s":18.368,"prefill_tps":95,"prompt_tokens":1624,"text":"The text describes the Nightshift runtime's process of scheduling inference jobs across device and cloud lanes, each with","ttft_s":17.102,"wall_s":18.376}],"seq":41,"t":116439,"type":"model_speed","was_cached":true}}
nightshift: [1792083018] page: {"bench":{"items":0,"model":"Phi-3.5-mini-instruct-q4f16_1-MLC","seconds":0,"seq":42,"t":116442,"type":"suite_done"}}
nightshift: [1792083031] page: {"bench":{"label":"load SmolLM2-1.7B-Instruct-q4f16_1-MLC","seq":43,"t":129719,"text":"Start to fetch params","type":"progress"}}
nightshift: [1792083031] page: {"bench":{"label":"load SmolLM2-1.7B-Instruct-q4f16_1-MLC","seq":44,"t":130020,"text":"Loading model from cache[3/37]: 96MB loaded. 10% completed, 13 secs elapsed.","type":"progress"}}
nightshift: [1792083032] page: {"bench":{"label":"load SmolLM2-1.7B-Instruct-q4f16_1-MLC","seq":45,"t":130361,"text":"Loading model from cache[7/37]: 199MB loaded. 21% completed, 14 secs elapsed.","type":"progress"}}
nightshift: [1792083032] page: {"bench":{"label":"load SmolLM2-1.7B-Instruct-q4f16_1-MLC","seq":46,"t":130744,"text":"Loading model from cache[11/37]: 298MB loaded. 32% completed, 14 secs elapsed.","type":"progress"}}
nightshift: [1792083032] page: {"bench":{"label":"load SmolLM2-1.7B-Instruct-q4f16_1-MLC","seq":47,"t":130915,"text":"Loading model from cache[14/37]: 370MB loaded. 40% completed, 14 secs elapsed.","type":"progress"}}
nightshift: [1792083033] page: {"bench":{"label":"load SmolLM2-1.7B-Instruct-q4f16_1-MLC","seq":48,"t":131226,"text":"Loading model from cache[19/37]: 487MB loaded. 52% completed, 15 secs elapsed.","type":"progress"}}
nightshift: [1792083033] page: {"bench":{"label":"load SmolLM2-1.7B-Instruct-q4f16_1-MLC","seq":49,"t":131404,"text":"Loading model from cache[22/37]: 559MB loaded. 60% completed, 15 secs elapsed.","type":"progress"}}
nightshift: [1792083033] page: {"bench":{"label":"load SmolLM2-1.7B-Instruct-q4f16_1-MLC","seq":50,"t":131637,"text":"Loading model from cache[26/37]: 658MB loaded. 71% completed, 15 secs elapsed.","type":"progress"}}
nightshift: [1792083033] page: {"bench":{"label":"load SmolLM2-1.7B-Instruct-q4f16_1-MLC","seq":51,"t":131872,"text":"Loading model from cache[30/37]: 746MB loaded. 81% completed, 15 secs elapsed.","type":"progress"}}
nightshift: [1792083033] page: {"bench":{"label":"load SmolLM2-1.7B-Instruct-q4f16_1-MLC","seq":52,"t":132143,"text":"Loading model from cache[34/37]: 847MB loaded. 92% completed, 15 secs elapsed.","type":"progress"}}
nightshift: [1792083034] page: {"bench":{"label":"load SmolLM2-1.7B-Instruct-q4f16_1-MLC","seq":53,"t":132341,"text":"Loading model from cache[37/37]: 919MB loaded. 100% completed, 16 secs elapsed.","type":"progress"}}
nightshift: [1792083043] page: {"bench":{"decode":[{"completion_tokens":15,"decode_tps":42.4,"e2e_s":0.641,"prefill_tps":227.8,"prompt_tokens":64,"text":"The bicycle, a mode of transportation that has evolved significantly over the centuries, has","ttft_s":0.281,"wall_s":0.646}],"first_call":{"ttft_s":0.451,"wall_s":0.595},"load_s":15.397,"mem_after_load":{"free_pct":44,"pressure":"normal","swap_used_bytes":2611085312},"mem_after_speed":{"free_pct":37,"pressure":"normal","swap_used_bytes":2611085312},"mem_before":{"free_pct":28,"pressure":"warn","swap_used_bytes":2611085312},"model":"SmolLM2-1.7B-Instruct-q4f16_1-MLC","prefill":[{"completion_tokens":23,"decode_tps":38.7,"e2e_s":8.041,"prefill_tps":204.4,"prompt_tokens":1521,"text":"The text above describes the architecture and functionality of the Nightshift runtime, including how it schedules inference jobs, caches weights","ttft_s":7.443,"wall_s":8.042}],"seq":54,"t":141648,"type":"model_speed","was_cached":true}}
nightshift: [1792083043] page: {"bench":{"items":0,"model":"SmolLM2-1.7B-Instruct-q4f16_1-MLC","seconds":0,"seq":55,"t":141657,"type":"suite_done"}}
nightshift: [1792083058] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":56,"t":156635,"text":"Start to fetch params","type":"progress"}}
nightshift: [1792083059] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":57,"t":157820,"text":"Loading model from cache[1/42]: 282MB loaded. 20% completed, 15 secs elapsed.","type":"progress"}}
nightshift: [1792083060] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":58,"t":158467,"text":"Loading model from cache[6/42]: 432MB loaded. 30% completed, 16 secs elapsed.","type":"progress"}}
nightshift: [1792083061] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":59,"t":159192,"text":"Loading model from cache[11/42]: 568MB loaded. 40% completed, 16 secs elapsed.","type":"progress"}}
nightshift: [1792083062] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":60,"t":160211,"text":"Loading model from cache[16/42]: 703MB loaded. 50% completed, 17 secs elapsed.","type":"progress"}}
nightshift: [1792083063] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":61,"t":161162,"text":"Loading model from cache[21/42]: 850MB loaded. 60% completed, 18 secs elapsed.","type":"progress"}}
nightshift: [1792083063] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":62,"t":161982,"text":"Loading model from cache[26/42]: 985MB loaded. 70% completed, 19 secs elapsed.","type":"progress"}}
nightshift: [1792083064] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":63,"t":162830,"text":"Loading model from cache[31/42]: 1131MB loaded. 80% completed, 20 secs elapsed.","type":"progress"}}
nightshift: [1792083065] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":64,"t":163575,"text":"Loading model from cache[36/42]: 1267MB loaded. 90% completed, 21 secs elapsed.","type":"progress"}}
nightshift: [1792083066] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":65,"t":164821,"text":"Loading model from cache[42/42]: 1403MB loaded. 100% completed, 22 secs elapsed.","type":"progress"}}
nightshift: [1792083081] page: {"bench":{"decode":[{"completion_tokens":15,"decode_tps":23,"e2e_s":1.038,"prefill_tps":112,"prompt_tokens":42,"text":"## The Pedal Revolution: A Century of Cycling\n\nThe bicycle, a seemingly simple","ttft_s":0.375,"wall_s":1.039}],"first_call":{"ttft_s":1.883,"wall_s":2.2},"load_s":22.701,"mem_after_load":{"free_pct":61,"pressure":"normal","swap_used_bytes":8253014016},"mem_after_speed":{"free_pct":55,"pressure":"normal","swap_used_bytes":6421282816},"mem_before":{"free_pct":37,"pressure":"normal","swap_used_bytes":2611085312},"model":"gemma-2-2b-it-q4f16_1-MLC","prefill":[{"completion_tokens":23,"decode_tps":17.6,"e2e_s":11.275,"prefill_tps":149.6,"prompt_tokens":1490,"text":"The text describes the architecture and workflow of the Nightshift runtime for inference jobs, emphasizing its use of device and cloud lanes","ttft_s":9.959,"wall_s":11.276}],"seq":66,"t":179396,"type":"model_speed","was_cached":true}}
nightshift: [1792083081] page: {"bench":{"items":0,"model":"gemma-2-2b-it-q4f16_1-MLC","seconds":0,"seq":67,"t":179411,"type":"suite_done"}}
nightshift: [1792083095] page: {"bench":{"label":"load Qwen2.5-Coder-3B-Instruct-q4f16_1-MLC","seq":68,"t":193891,"text":"Start to fetch params","type":"progress"}}
nightshift: [1792083096] page: {"bench":{"label":"load Qwen2.5-Coder-3B-Instruct-q4f16_1-MLC","seq":69,"t":194493,"text":"Loading model from cache[2/62]: 170MB loaded. 10% completed, 14 secs elapsed.","type":"progress"}}
nightshift: [1792083096] page: {"bench":{"label":"load Qwen2.5-Coder-3B-Instruct-q4f16_1-MLC","seq":70,"t":194923,"text":"Loading model from cache[9/62]: 347MB loaded. 20% completed, 15 secs elapsed.","type":"progress"}}
nightshift: [1792083097] page: {"bench":{"label":"load Qwen2.5-Coder-3B-Instruct-q4f16_1-MLC","seq":71,"t":195264,"text":"Loading model from cache[15/62]: 498MB loaded. 30% completed, 15 secs elapsed.","type":"progress"}}
nightshift: [1792083097] page: {"bench":{"label":"load Qwen2.5-Coder-3B-Instruct-q4f16_1-MLC","seq":72,"t":195635,"text":"Loading model from cache[22/62]: 665MB loaded. 40% completed, 15 secs elapsed.","type":"progress"}}
nightshift: [1792083097] page: {"bench":{"label":"load Qwen2.5-Coder-3B-Instruct-q4f16_1-MLC","seq":73,"t":195995,"text":"Loading model from cache[29/62]: 843MB loaded. 50% completed, 16 secs elapsed.","type":"progress"}}
nightshift: [1792083098] page: {"bench":{"label":"load Qwen2.5-Coder-3B-Instruct-q4f16_1-MLC","seq":74,"t":196413,"text":"Loading model from cache[35/62]: 995MB loaded. 60% completed, 16 secs elapsed.","type":"progress"}}
nightshift: [1792083098] page: {"bench":{"label":"load Qwen2.5-Coder-3B-Instruct-q4f16_1-MLC","seq":75,"t":196845,"text":"Loading model from cache[42/62]: 1165MB loaded. 70% completed, 16 secs elapsed.","type":"progress"}}
nightshift: [1792083099] page: {"bench":{"label":"load Qwen2.5-Coder-3B-Instruct-q4f16_1-MLC","seq":76,"t":197239,"text":"Loading model from cache[49/62]: 1332MB loaded. 80% completed, 17 secs elapsed.","type":"progress"}}
nightshift: [1792083099] page: {"bench":{"label":"load Qwen2.5-Coder-3B-Instruct-q4f16_1-MLC","seq":77,"t":197670,"text":"Loading model from cache[56/62]: 1505MB loaded. 90% completed, 17 secs elapsed.","type":"progress"}}
nightshift: [1792083099] page: {"bench":{"label":"load Qwen2.5-Coder-3B-Instruct-q4f16_1-MLC","seq":78,"t":198057,"text":"Loading model from cache[62/62]: 1656MB loaded. 100% completed, 18 secs elapsed.","type":"progress"}}
nightshift: [1792083118] page: {"bench":{"decode":[{"completion_tokens":15,"decode_tps":18.6,"e2e_s":1.362,"prefill_tps":96.4,"prompt_tokens":53,"text":"The bicycle has a rich and fascinating history that spans over two centuries. The first","ttft_s":0.55,"wall_s":1.363}],"first_call":{"ttft_s":0.486,"wall_s":0.775},"load_s":17.991,"mem_after_load":{"free_pct":61,"pressure":"normal","swap_used_bytes":5842206720},"mem_after_speed":{"free_pct":58,"pressure":"normal","swap_used_bytes":5135138816},"mem_before":{"free_pct":64,"pressure":"normal","swap_used_bytes":6354173952},"model":"Qwen2.5-Coder-3B-Instruct-q4f16_1-MLC","prefill":[{"completion_tokens":23,"decode_tps":14.1,"e2e_s":16.234,"prefill_tps":101.8,"prompt_tokens":1487,"text":"The text above describes the Nightshift runtime, which schedules inference jobs across device and cloud lanes, with each lane being a","ttft_s":14.601,"wall_s":16.236}],"seq":79,"t":216462,"type":"model_speed","was_cached":true}}
nightshift: [1792083118] page: {"bench":{"items":0,"model":"Qwen2.5-Coder-3B-Instruct-q4f16_1-MLC","seconds":0,"seq":80,"t":216483,"type":"suite_done"}}
nightshift: [1792083118] page: {"bench":{"phase":"models","seconds":213.497,"seq":81,"t":216488,"type":"phase_end"}}
nightshift: [1792083118] page: {"bench":{"errors":[],"seconds":213.59,"seq":82,"t":216512,"type":"done"}}
