nightshift: [1792087183] start v0.90.0 · Apple M5 (m5+) · 16 GB · macOS 26.4 · 4P+6E cores
nightshift: [1792087184] serving /Users/samuelcummings/nightshift-m1m4-build/web-bench on http://127.0.0.1:47321/
nightshift: [1792087184] webkit: BackgroundWebContentRunningBoardThrottlingEnabled=false, HiddenPageDOMTimerThrottlingAutoIncreases=false, HiddenPageDOMTimerThrottlingEnabled=false, PageVisibilityBasedProcessSuppressionEnabled=false, WebGPUEnabled=true
nightshift: [1792087184] webkit: window occlusion detection off (page stays active behind other windows): true
nightshift: [1792087184] cockpit loading (injected scripts removed: true)
nightshift: [1792087184] quit gate installed (applicationShouldTerminate:)
nightshift: [1792087187] page: {"bench":{"label":"97-loadmem","phases":["env","models"],"seq":0,"t":2820,"type":"start"}}
nightshift: [1792087187] page: {"bench":{"phase":"env","seq":1,"t":2822,"type":"phase_start"}}
nightshift: [1792087187] page: {"bench":{"boot":{"bench_script_ms":154,"dom_content_loaded_ms":268,"engine_ready_ms":179,"load_event_ms":318,"resources":211,"scripts":175,"shell_mounted_ms":322,"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":60,"pressure":"normal","swap_used_bytes":4863492096},"seq":2,"storage":{"persisted":false,"quota_gb":41.2,"usage_mb":11530},"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":2876,"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":0.05}}}
nightshift: [1792087187] page: {"bench":{"phase":"env","seconds":0.059,"seq":3,"t":2881,"type":"phase_end"}}
nightshift: [1792087187] page: {"bench":{"phase":"models","seq":4,"t":2882,"type":"phase_start"}}
nightshift: [1792087192] 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":11530},"tauri_globals":false,"tier":"16 GB","webllm":{"prebuilt":163,"worker_webgpu":true}}}
nightshift: [1792087203] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":5,"t":19523,"text":"Start to fetch params","type":"progress"}}
nightshift: [1792087205] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":6,"t":20634,"text":"Loading model from cache[1/42]: 282MB loaded. 20% completed, 17 secs elapsed.","type":"progress"}}
nightshift: [1792087205] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":7,"t":21153,"text":"Loading model from cache[6/42]: 432MB loaded. 30% completed, 17 secs elapsed.","type":"progress"}}
nightshift: [1792087206] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":8,"t":21880,"text":"Loading model from cache[11/42]: 568MB loaded. 40% completed, 18 secs elapsed.","type":"progress"}}
nightshift: [1792087206] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":9,"t":22387,"text":"Loading model from cache[16/42]: 703MB loaded. 50% completed, 19 secs elapsed.","type":"progress"}}
nightshift: [1792087207] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":10,"t":23021,"text":"Loading model from cache[21/42]: 850MB loaded. 60% completed, 19 secs elapsed.","type":"progress"}}
nightshift: [1792087207] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":11,"t":23507,"text":"Loading model from cache[26/42]: 985MB loaded. 70% completed, 20 secs elapsed.","type":"progress"}}
nightshift: [1792087208] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":12,"t":23963,"text":"Loading model from cache[31/42]: 1131MB loaded. 80% completed, 20 secs elapsed.","type":"progress"}}
nightshift: [1792087208] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":13,"t":24381,"text":"Loading model from cache[36/42]: 1267MB loaded. 90% completed, 21 secs elapsed.","type":"progress"}}
nightshift: [1792087209] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":14,"t":24719,"text":"Loading model from cache[42/42]: 1403MB loaded. 100% completed, 21 secs elapsed.","type":"progress"}}
nightshift: [1792087222] page: {"bench":{"decode":[{"completion_tokens":15,"decode_tps":29.1,"e2e_s":0.905,"prefill_tps":108.2,"prompt_tokens":42,"text":"## The Pedal Revolution: A Century of Cycling\n\nThe bicycle, a seemingly simple","ttft_s":0.388,"wall_s":0.905}],"first_call":{"ttft_s":0.521,"wall_s":0.752},"load_s":21.355,"mem_after_load":{"free_pct":40,"pressure":"warn","swap_used_bytes":5926617088},"mem_after_speed":{"free_pct":40,"pressure":"normal","swap_used_bytes":6087770112},"mem_before":{"free_pct":60,"pressure":"normal","swap_used_bytes":4863492096},"model":"gemma-2-2b-it-q4f16_1-MLC","prefill":[{"completion_tokens":23,"decode_tps":19.3,"e2e_s":11.979,"prefill_tps":138.2,"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":10.782,"wall_s":11.982}],"seq":15,"t":38383,"type":"model_speed","was_cached":true}}
nightshift: [1792087222] page: {"bench":{"items":0,"model":"gemma-2-2b-it-q4f16_1-MLC","seconds":0,"seq":16,"t":38387,"type":"suite_done"}}
nightshift: [1792087237] page: {"bench":{"label":"load Llama-3.2-3B-Instruct-q4f16_1-MLC","seq":17,"t":53306,"text":"Start to fetch params","type":"progress"}}
nightshift: [1792087238] page: {"bench":{"label":"load Llama-3.2-3B-Instruct-q4f16_1-MLC","seq":18,"t":54185,"text":"Loading model from cache[1/58]: 188MB loaded. 10% completed, 15 secs elapsed.","type":"progress"}}
nightshift: [1792087239] page: {"bench":{"label":"load Llama-3.2-3B-Instruct-q4f16_1-MLC","seq":19,"t":54790,"text":"Loading model from cache[8/58]: 374MB loaded. 21% completed, 16 secs elapsed.","type":"progress"}}
nightshift: [1792087239] page: {"bench":{"label":"load Llama-3.2-3B-Instruct-q4f16_1-MLC","seq":20,"t":55344,"text":"Loading model from cache[14/58]: 536MB loaded. 31% completed, 17 secs elapsed.","type":"progress"}}
nightshift: [1792087240] page: {"bench":{"label":"load Llama-3.2-3B-Instruct-q4f16_1-MLC","seq":21,"t":56036,"text":"Loading model from cache[20/58]: 698MB loaded. 40% completed, 17 secs elapsed.","type":"progress"}}
nightshift: [1792087240] page: {"bench":{"label":"load Llama-3.2-3B-Instruct-q4f16_1-MLC","seq":22,"t":56531,"text":"Loading model from cache[27/58]: 884MB loaded. 51% completed, 18 secs elapsed.","type":"progress"}}
nightshift: [1792087241] page: {"bench":{"label":"load Llama-3.2-3B-Instruct-q4f16_1-MLC","seq":23,"t":56841,"text":"Loading model from cache[33/58]: 1046MB loaded. 60% completed, 18 secs elapsed.","type":"progress"}}
nightshift: [1792087241] page: {"bench":{"label":"load Llama-3.2-3B-Instruct-q4f16_1-MLC","seq":24,"t":57157,"text":"Loading model from cache[39/58]: 1208MB loaded. 70% completed, 18 secs elapsed.","type":"progress"}}
nightshift: [1792087241] page: {"bench":{"label":"load Llama-3.2-3B-Instruct-q4f16_1-MLC","seq":25,"t":57521,"text":"Loading model from cache[46/58]: 1400MB loaded. 81% completed, 19 secs elapsed.","type":"progress"}}
nightshift: [1792087242] page: {"bench":{"label":"load Llama-3.2-3B-Instruct-q4f16_1-MLC","seq":26,"t":57822,"text":"Loading model from cache[52/58]: 1562MB loaded. 90% completed, 19 secs elapsed.","type":"progress"}}
nightshift: [1792087242] page: {"bench":{"label":"load Llama-3.2-3B-Instruct-q4f16_1-MLC","seq":27,"t":58135,"text":"Loading model from cache[58/58]: 1724MB loaded. 100% completed, 19 secs elapsed.","type":"progress"}}
nightshift: [1792087259] page: {"bench":{"decode":[{"completion_tokens":15,"decode_tps":24.2,"e2e_s":1.208,"prefill_tps":92.5,"prompt_tokens":54,"text":"The bicycle, a ubiquitous mode of transportation and recreation, has a rich and fascinating","ttft_s":0.584,"wall_s":1.209}],"first_call":{"ttft_s":0.452,"wall_s":0.747},"load_s":19.242,"mem_after_load":{"free_pct":57,"pressure":"normal","swap_used_bytes":7335116800},"mem_after_speed":{"free_pct":46,"pressure":"normal","swap_used_bytes":6915555328},"mem_before":{"free_pct":39,"pressure":"normal","swap_used_bytes":6087770112},"model":"Llama-3.2-3B-Instruct-q4f16_1-MLC","prefill":[{"completion_tokens":23,"decode_tps":17.1,"e2e_s":14.843,"prefill_tps":108.1,"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":13.493,"wall_s":14.845}],"seq":28,"t":74954,"type":"model_speed","was_cached":true}}
nightshift: [1792087259] page: {"bench":{"items":0,"model":"Llama-3.2-3B-Instruct-q4f16_1-MLC","seconds":0,"seq":29,"t":74956,"type":"suite_done"}}
nightshift: [1792087273] page: {"bench":{"label":"load Phi-3.5-mini-instruct-q4f16_1-MLC","seq":30,"t":88763,"text":"Start to fetch params","type":"progress"}}
nightshift: [1792087273] page: {"bench":{"label":"load Phi-3.5-mini-instruct-q4f16_1-MLC","seq":31,"t":89406,"text":"Loading model from cache[8/83]: 215MB loaded. 10% completed, 14 secs elapsed.","type":"progress"}}
nightshift: [1792087274] page: {"bench":{"label":"load Phi-3.5-mini-instruct-q4f16_1-MLC","seq":32,"t":89886,"text":"Loading model from cache[16/83]: 413MB loaded. 20% completed, 15 secs elapsed.","type":"progress"}}
nightshift: [1792087274] page: {"bench":{"label":"load Phi-3.5-mini-instruct-q4f16_1-MLC","seq":33,"t":90372,"text":"Loading model from cache[25/83]: 636MB loaded. 30% completed, 15 secs elapsed.","type":"progress"}}
nightshift: [1792087275] page: {"bench":{"label":"load Phi-3.5-mini-instruct-q4f16_1-MLC","seq":34,"t":90793,"text":"Loading model from cache[33/83]: 839MB loaded. 40% completed, 16 secs elapsed.","type":"progress"}}
nightshift: [1792087275] page: {"bench":{"label":"load Phi-3.5-mini-instruct-q4f16_1-MLC","seq":35,"t":91214,"text":"Loading model from cache[41/83]: 1037MB loaded. 50% completed, 16 secs elapsed.","type":"progress"}}
nightshift: [1792087276] page: {"bench":{"label":"load Phi-3.5-mini-instruct-q4f16_1-MLC","seq":36,"t":91754,"text":"Loading model from cache[49/83]: 1235MB loaded. 60% completed, 17 secs elapsed.","type":"progress"}}
nightshift: [1792087276] page: {"bench":{"label":"load Phi-3.5-mini-instruct-q4f16_1-MLC","seq":37,"t":92192,"text":"Loading model from cache[58/83]: 1447MB loaded. 70% completed, 17 secs elapsed.","type":"progress"}}
nightshift: [1792087277] page: {"bench":{"label":"load Phi-3.5-mini-instruct-q4f16_1-MLC","seq":38,"t":92649,"text":"Loading model from cache[66/83]: 1649MB loaded. 80% completed, 18 secs elapsed.","type":"progress"}}
nightshift: [1792087277] page: {"bench":{"label":"load Phi-3.5-mini-instruct-q4f16_1-MLC","seq":39,"t":93118,"text":"Loading model from cache[74/83]: 1848MB loaded. 90% completed, 18 secs elapsed.","type":"progress"}}
nightshift: [1792087278] page: {"bench":{"label":"load Phi-3.5-mini-instruct-q4f16_1-MLC","seq":40,"t":93654,"text":"Loading model from cache[83/83]: 2051MB loaded. 100% completed, 19 secs elapsed.","type":"progress"}}
nightshift: [1792087298] page: {"bench":{"decode":[{"completion_tokens":15,"decode_tps":24.5,"e2e_s":1.285,"prefill_tps":93,"prompt_tokens":62,"text":"The bicycle, a marvel of human ingenuity, has under","ttft_s":0.667,"wall_s":1.285}],"first_call":{"ttft_s":0.583,"wall_s":0.863},"load_s":18.205,"mem_after_load":{"free_pct":50,"pressure":"normal","swap_used_bytes":6688997376},"mem_after_speed":{"free_pct":40,"pressure":"normal","swap_used_bytes":6529613824},"mem_before":{"free_pct":49,"pressure":"normal","swap_used_bytes":6898778112},"model":"Phi-3.5-mini-instruct-q4f16_1-MLC","prefill":[{"completion_tokens":23,"decode_tps":21,"e2e_s":18.207,"prefill_tps":94.9,"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":17.11,"wall_s":18.209}],"seq":41,"t":114036,"type":"model_speed","was_cached":true}}
nightshift: [1792087298] page: {"bench":{"items":0,"model":"Phi-3.5-mini-instruct-q4f16_1-MLC","seconds":0,"seq":42,"t":114038,"type":"suite_done"}}
nightshift: [1792087312] page: {"bench":{"label":"load SmolLM2-1.7B-Instruct-q4f16_1-MLC","seq":43,"t":127679,"text":"Start to fetch params","type":"progress"}}
nightshift: [1792087312] page: {"bench":{"label":"load SmolLM2-1.7B-Instruct-q4f16_1-MLC","seq":44,"t":127943,"text":"Loading model from cache[3/37]: 96MB loaded. 10% completed, 14 secs elapsed.","type":"progress"}}
nightshift: [1792087312] page: {"bench":{"label":"load SmolLM2-1.7B-Instruct-q4f16_1-MLC","seq":45,"t":128179,"text":"Loading model from cache[7/37]: 199MB loaded. 21% completed, 14 secs elapsed.","type":"progress"}}
nightshift: [1792087312] page: {"bench":{"label":"load SmolLM2-1.7B-Instruct-q4f16_1-MLC","seq":46,"t":128500,"text":"Loading model from cache[11/37]: 298MB loaded. 32% completed, 14 secs elapsed.","type":"progress"}}
nightshift: [1792087313] page: {"bench":{"label":"load SmolLM2-1.7B-Instruct-q4f16_1-MLC","seq":47,"t":128681,"text":"Loading model from cache[14/37]: 370MB loaded. 40% completed, 14 secs elapsed.","type":"progress"}}
nightshift: [1792087313] page: {"bench":{"label":"load SmolLM2-1.7B-Instruct-q4f16_1-MLC","seq":48,"t":129082,"text":"Loading model from cache[19/37]: 487MB loaded. 52% completed, 15 secs elapsed.","type":"progress"}}
nightshift: [1792087313] page: {"bench":{"label":"load SmolLM2-1.7B-Instruct-q4f16_1-MLC","seq":49,"t":129313,"text":"Loading model from cache[22/37]: 559MB loaded. 60% completed, 15 secs elapsed.","type":"progress"}}
nightshift: [1792087314] page: {"bench":{"label":"load SmolLM2-1.7B-Instruct-q4f16_1-MLC","seq":50,"t":129636,"text":"Loading model from cache[26/37]: 658MB loaded. 71% completed, 15 secs elapsed.","type":"progress"}}
nightshift: [1792087314] page: {"bench":{"label":"load SmolLM2-1.7B-Instruct-q4f16_1-MLC","seq":51,"t":129930,"text":"Loading model from cache[30/37]: 746MB loaded. 81% completed, 16 secs elapsed.","type":"progress"}}
nightshift: [1792087314] page: {"bench":{"label":"load SmolLM2-1.7B-Instruct-q4f16_1-MLC","seq":52,"t":130272,"text":"Loading model from cache[34/37]: 847MB loaded. 92% completed, 16 secs elapsed.","type":"progress"}}
nightshift: [1792087314] page: {"bench":{"label":"load SmolLM2-1.7B-Instruct-q4f16_1-MLC","seq":53,"t":130512,"text":"Loading model from cache[37/37]: 919MB loaded. 100% completed, 16 secs elapsed.","type":"progress"}}
nightshift: [1792087324] page: {"bench":{"decode":[{"completion_tokens":15,"decode_tps":31.6,"e2e_s":0.752,"prefill_tps":233.6,"prompt_tokens":64,"text":"The bicycle, a mode of transportation that has evolved significantly over the centuries, has","ttft_s":0.274,"wall_s":0.752}],"first_call":{"ttft_s":0.442,"wall_s":0.619},"load_s":15.989,"mem_after_load":{"free_pct":54,"pressure":"normal","swap_used_bytes":6236012544},"mem_after_speed":{"free_pct":49,"pressure":"normal","swap_used_bytes":6210846720},"mem_before":{"free_pct":41,"pressure":"normal","swap_used_bytes":6529613824},"model":"SmolLM2-1.7B-Instruct-q4f16_1-MLC","prefill":[{"completion_tokens":23,"decode_tps":38.9,"e2e_s":8.449,"prefill_tps":193.7,"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":7.852,"wall_s":8.449}],"seq":54,"t":140366,"type":"model_speed","was_cached":true}}
nightshift: [1792087324] page: {"bench":{"items":0,"model":"SmolLM2-1.7B-Instruct-q4f16_1-MLC","seconds":0,"seq":55,"t":140371,"type":"suite_done"}}
nightshift: [1792087339] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":56,"t":155377,"text":"Start to fetch params","type":"progress"}}
nightshift: [1792087340] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":57,"t":156468,"text":"Loading model from cache[1/42]: 282MB loaded. 20% completed, 15 secs elapsed.","type":"progress"}}
nightshift: [1792087341] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":58,"t":156824,"text":"Loading model from cache[6/42]: 432MB loaded. 30% completed, 15 secs elapsed.","type":"progress"}}
nightshift: [1792087341] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":59,"t":157145,"text":"Loading model from cache[11/42]: 568MB loaded. 40% completed, 16 secs elapsed.","type":"progress"}}
nightshift: [1792087341] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":60,"t":157558,"text":"Loading model from cache[16/42]: 703MB loaded. 50% completed, 16 secs elapsed.","type":"progress"}}
nightshift: [1792087342] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":61,"t":158002,"text":"Loading model from cache[21/42]: 850MB loaded. 60% completed, 17 secs elapsed.","type":"progress"}}
nightshift: [1792087342] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":62,"t":158371,"text":"Loading model from cache[26/42]: 985MB loaded. 70% completed, 17 secs elapsed.","type":"progress"}}
nightshift: [1792087343] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":63,"t":158734,"text":"Loading model from cache[31/42]: 1131MB loaded. 80% completed, 17 secs elapsed.","type":"progress"}}
nightshift: [1792087343] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":64,"t":159076,"text":"Loading model from cache[36/42]: 1267MB loaded. 90% completed, 18 secs elapsed.","type":"progress"}}
nightshift: [1792087343] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":65,"t":159470,"text":"Loading model from cache[42/42]: 1403MB loaded. 100% completed, 18 secs elapsed.","type":"progress"}}
nightshift: [1792087358] page: {"bench":{"decode":[{"completion_tokens":15,"decode_tps":19.2,"e2e_s":1.169,"prefill_tps":109.4,"prompt_tokens":42,"text":"## The Pedal Revolution: A Century of Cycling\n\nThe bicycle, a seemingly simple","ttft_s":0.384,"wall_s":1.17}],"first_call":{"ttft_s":0.379,"wall_s":0.586},"load_s":18.606,"mem_after_load":{"free_pct":50,"pressure":"normal","swap_used_bytes":6009389056},"mem_after_speed":{"free_pct":42,"pressure":"normal","swap_used_bytes":5874974720},"mem_before":{"free_pct":50,"pressure":"normal","swap_used_bytes":6210846720},"model":"gemma-2-2b-it-q4f16_1-MLC","prefill":[{"completion_tokens":23,"decode_tps":18.9,"e2e_s":12.401,"prefill_tps":133.3,"prompt_tokens":1490,"text":"The text describes the architecture and workflow of the Nightshift runtime for running inference jobs across device and cloud lanes. \n","ttft_s":11.178,"wall_s":12.401}],"seq":66,"t":173647,"type":"model_speed","was_cached":true}}
nightshift: [1792087358] page: {"bench":{"items":0,"model":"gemma-2-2b-it-q4f16_1-MLC","seconds":0,"seq":67,"t":173652,"type":"suite_done"}}
nightshift: [1792087372] page: {"bench":{"label":"load Qwen2.5-Coder-3B-Instruct-q4f16_1-MLC","seq":68,"t":187918,"text":"Start to fetch params","type":"progress"}}
nightshift: [1792087372] page: {"bench":{"label":"load Qwen2.5-Coder-3B-Instruct-q4f16_1-MLC","seq":69,"t":188478,"text":"Loading model from cache[2/62]: 170MB loaded. 10% completed, 14 secs elapsed.","type":"progress"}}
nightshift: [1792087373] page: {"bench":{"label":"load Qwen2.5-Coder-3B-Instruct-q4f16_1-MLC","seq":70,"t":189024,"text":"Loading model from cache[9/62]: 347MB loaded. 20% completed, 15 secs elapsed.","type":"progress"}}
nightshift: [1792087373] page: {"bench":{"label":"load Qwen2.5-Coder-3B-Instruct-q4f16_1-MLC","seq":71,"t":189543,"text":"Loading model from cache[15/62]: 498MB loaded. 30% completed, 15 secs elapsed.","type":"progress"}}
nightshift: [1792087374] page: {"bench":{"label":"load Qwen2.5-Coder-3B-Instruct-q4f16_1-MLC","seq":72,"t":190092,"text":"Loading model from cache[22/62]: 665MB loaded. 40% completed, 16 secs elapsed.","type":"progress"}}
nightshift: [1792087374] page: {"bench":{"label":"load Qwen2.5-Coder-3B-Instruct-q4f16_1-MLC","seq":73,"t":190607,"text":"Loading model from cache[29/62]: 843MB loaded. 50% completed, 16 secs elapsed.","type":"progress"}}
nightshift: [1792087375] page: {"bench":{"label":"load Qwen2.5-Coder-3B-Instruct-q4f16_1-MLC","seq":74,"t":191109,"text":"Loading model from cache[35/62]: 995MB loaded. 60% completed, 17 secs elapsed.","type":"progress"}}
nightshift: [1792087376] page: {"bench":{"label":"load Qwen2.5-Coder-3B-Instruct-q4f16_1-MLC","seq":75,"t":191653,"text":"Loading model from cache[42/62]: 1165MB loaded. 70% completed, 18 secs elapsed.","type":"progress"}}
nightshift: [1792087376] page: {"bench":{"label":"load Qwen2.5-Coder-3B-Instruct-q4f16_1-MLC","seq":76,"t":192220,"text":"Loading model from cache[49/62]: 1332MB loaded. 80% completed, 18 secs elapsed.","type":"progress"}}
nightshift: [1792087377] page: {"bench":{"label":"load Qwen2.5-Coder-3B-Instruct-q4f16_1-MLC","seq":77,"t":192720,"text":"Loading model from cache[56/62]: 1505MB loaded. 90% completed, 19 secs elapsed.","type":"progress"}}
nightshift: [1792087377] page: {"bench":{"label":"load Qwen2.5-Coder-3B-Instruct-q4f16_1-MLC","seq":78,"t":193119,"text":"Loading model from cache[62/62]: 1656MB loaded. 100% completed, 19 secs elapsed.","type":"progress"}}
nightshift: [1792087396] page: {"bench":{"decode":[{"completion_tokens":15,"decode_tps":14.5,"e2e_s":1.639,"prefill_tps":88.9,"prompt_tokens":53,"text":"The bicycle has a rich and fascinating history that spans over two centuries. The first","ttft_s":0.596,"wall_s":1.64}],"first_call":{"ttft_s":0.51,"wall_s":0.782},"load_s":18.944,"mem_after_load":{"free_pct":51,"pressure":"normal","swap_used_bytes":5858197504},"mem_after_speed":{"free_pct":43,"pressure":"normal","swap_used_bytes":5791088640},"mem_before":{"free_pct":42,"pressure":"normal","swap_used_bytes":5874974720},"model":"Qwen2.5-Coder-3B-Instruct-q4f16_1-MLC","prefill":[{"completion_tokens":23,"decode_tps":13.1,"e2e_s":16.596,"prefill_tps":100.3,"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":14.832,"wall_s":16.603}],"seq":79,"t":212173,"type":"model_speed","was_cached":true}}
nightshift: [1792087396] page: {"bench":{"items":0,"model":"Qwen2.5-Coder-3B-Instruct-q4f16_1-MLC","seconds":0,"seq":80,"t":212176,"type":"suite_done"}}
nightshift: [1792087396] page: {"bench":{"phase":"models","seconds":209.298,"seq":81,"t":212180,"type":"phase_end"}}
nightshift: [1792087396] page: {"bench":{"errors":[],"seconds":209.36,"seq":82,"t":212182,"type":"done"}}
