nightshift: [1792084221] start v0.90.0 · Apple M5 (m5+) · 16 GB · macOS 26.4 · 4P+6E cores
nightshift: [1792084221] serving /Users/samuelcummings/nightshift-m1m4-build/web-bench on http://127.0.0.1:47321/
nightshift: [1792084221] webkit: BackgroundWebContentRunningBoardThrottlingEnabled=false, HiddenPageDOMTimerThrottlingAutoIncreases=false, HiddenPageDOMTimerThrottlingEnabled=false, PageVisibilityBasedProcessSuppressionEnabled=false, WebGPUEnabled=true
nightshift: [1792084221] webkit: window occlusion detection off (page stays active behind other windows): true
nightshift: [1792084221] cockpit loading (injected scripts removed: true)
nightshift: [1792084221] quit gate installed (applicationShouldTerminate:)
nightshift: [1792084224] page: {"bench":{"label":"97-loadmem","phases":["env","models"],"seq":0,"t":2876,"type":"start"}}
nightshift: [1792084224] page: {"bench":{"phase":"env","seq":1,"t":2878,"type":"phase_start"}}
nightshift: [1792084225] page: {"bench":{"boot":{"bench_script_ms":190,"dom_content_loaded_ms":317,"engine_ready_ms":219,"load_event_ms":373,"resources":211,"scripts":175,"shell_mounted_ms":379,"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":64,"pressure":"normal","swap_used_bytes":5983305728},"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":2964,"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":0.08}}}
nightshift: [1792084225] page: {"bench":{"phase":"env","seconds":0.093,"seq":3,"t":2971,"type":"phase_end"}}
nightshift: [1792084225] page: {"bench":{"phase":"models","seq":4,"t":2974,"type":"phase_start"}}
nightshift: [1792084230] 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: [1792084244] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":5,"t":22137,"text":"Start to fetch params","type":"progress"}}
nightshift: [1792084245] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":6,"t":23259,"text":"Loading model from cache[1/42]: 282MB loaded. 20% completed, 19 secs elapsed.","type":"progress"}}
nightshift: [1792084245] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":7,"t":23727,"text":"Loading model from cache[6/42]: 432MB loaded. 30% completed, 19 secs elapsed.","type":"progress"}}
nightshift: [1792084246] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":8,"t":24419,"text":"Loading model from cache[11/42]: 568MB loaded. 40% completed, 20 secs elapsed.","type":"progress"}}
nightshift: [1792084247] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":9,"t":25041,"text":"Loading model from cache[16/42]: 703MB loaded. 50% completed, 20 secs elapsed.","type":"progress"}}
nightshift: [1792084247] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":10,"t":25741,"text":"Loading model from cache[21/42]: 850MB loaded. 60% completed, 21 secs elapsed.","type":"progress"}}
nightshift: [1792084248] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":11,"t":26146,"text":"Loading model from cache[26/42]: 985MB loaded. 70% completed, 22 secs elapsed.","type":"progress"}}
nightshift: [1792084248] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":12,"t":26678,"text":"Loading model from cache[31/42]: 1131MB loaded. 80% completed, 22 secs elapsed.","type":"progress"}}
nightshift: [1792084249] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":13,"t":27281,"text":"Loading model from cache[36/42]: 1267MB loaded. 90% completed, 23 secs elapsed.","type":"progress"}}
nightshift: [1792084249] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":14,"t":27824,"text":"Loading model from cache[42/42]: 1403MB loaded. 100% completed, 23 secs elapsed.","type":"progress"}}
nightshift: [1792084264] page: {"bench":{"decode":[{"completion_tokens":15,"decode_tps":22.6,"e2e_s":1.026,"prefill_tps":118.3,"prompt_tokens":42,"text":"## The Pedal Revolution: A Century of Cycling\n\nThe bicycle, a seemingly simple","ttft_s":0.355,"wall_s":1.026}],"first_call":{"ttft_s":0.574,"wall_s":0.826},"load_s":24.375,"mem_after_load":{"free_pct":33,"pressure":"warn","swap_used_bytes":6547177472},"mem_after_speed":{"free_pct":37,"pressure":"normal","swap_used_bytes":6997540864},"mem_before":{"free_pct":64,"pressure":"normal","swap_used_bytes":5983305728},"model":"gemma-2-2b-it-q4f16_1-MLC","prefill":[{"completion_tokens":23,"decode_tps":20.2,"e2e_s":12.552,"prefill_tps":130.6,"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":11.405,"wall_s":12.557}],"seq":15,"t":42275,"type":"model_speed","was_cached":true}}
nightshift: [1792084264] page: {"bench":{"items":0,"model":"gemma-2-2b-it-q4f16_1-MLC","seconds":0,"seq":16,"t":42278,"type":"suite_done"}}
nightshift: [1792084279] page: {"bench":{"label":"load Llama-3.2-3B-Instruct-q4f16_1-MLC","seq":17,"t":57725,"text":"Start to fetch params","type":"progress"}}
nightshift: [1792084280] page: {"bench":{"label":"load Llama-3.2-3B-Instruct-q4f16_1-MLC","seq":18,"t":58512,"text":"Loading model from cache[1/58]: 188MB loaded. 10% completed, 16 secs elapsed.","type":"progress"}}
nightshift: [1792084281] page: {"bench":{"label":"load Llama-3.2-3B-Instruct-q4f16_1-MLC","seq":19,"t":58971,"text":"Loading model from cache[8/58]: 374MB loaded. 21% completed, 16 secs elapsed.","type":"progress"}}
nightshift: [1792084281] page: {"bench":{"label":"load Llama-3.2-3B-Instruct-q4f16_1-MLC","seq":20,"t":59741,"text":"Loading model from cache[14/58]: 536MB loaded. 31% completed, 17 secs elapsed.","type":"progress"}}
nightshift: [1792084282] page: {"bench":{"label":"load Llama-3.2-3B-Instruct-q4f16_1-MLC","seq":21,"t":60588,"text":"Loading model from cache[20/58]: 698MB loaded. 40% completed, 18 secs elapsed.","type":"progress"}}
nightshift: [1792084283] page: {"bench":{"label":"load Llama-3.2-3B-Instruct-q4f16_1-MLC","seq":22,"t":61048,"text":"Loading model from cache[27/58]: 884MB loaded. 51% completed, 18 secs elapsed.","type":"progress"}}
nightshift: [1792084283] page: {"bench":{"label":"load Llama-3.2-3B-Instruct-q4f16_1-MLC","seq":23,"t":61495,"text":"Loading model from cache[33/58]: 1046MB loaded. 60% completed, 18 secs elapsed.","type":"progress"}}
nightshift: [1792084284] page: {"bench":{"label":"load Llama-3.2-3B-Instruct-q4f16_1-MLC","seq":24,"t":62252,"text":"Loading model from cache[39/58]: 1208MB loaded. 70% completed, 19 secs elapsed.","type":"progress"}}
nightshift: [1792084284] page: {"bench":{"label":"load Llama-3.2-3B-Instruct-q4f16_1-MLC","seq":25,"t":62896,"text":"Loading model from cache[46/58]: 1400MB loaded. 81% completed, 20 secs elapsed.","type":"progress"}}
nightshift: [1792084285] page: {"bench":{"label":"load Llama-3.2-3B-Instruct-q4f16_1-MLC","seq":26,"t":63241,"text":"Loading model from cache[52/58]: 1562MB loaded. 90% completed, 20 secs elapsed.","type":"progress"}}
nightshift: [1792084285] page: {"bench":{"label":"load Llama-3.2-3B-Instruct-q4f16_1-MLC","seq":27,"t":63579,"text":"Loading model from cache[58/58]: 1724MB loaded. 100% completed, 21 secs elapsed.","type":"progress"}}
nightshift: [1792084302] page: {"bench":{"decode":[{"completion_tokens":15,"decode_tps":21.7,"e2e_s":1.234,"prefill_tps":100,"prompt_tokens":54,"text":"The bicycle, a ubiquitous mode of transportation and recreation, has a rich and fascinating","ttft_s":0.54,"wall_s":1.241}],"first_call":{"ttft_s":0.413,"wall_s":0.654},"load_s":20.784,"mem_after_load":{"free_pct":49,"pressure":"normal","swap_used_bytes":7674396672},"mem_after_speed":{"free_pct":48,"pressure":"normal","swap_used_bytes":7607222272},"mem_before":{"free_pct":42,"pressure":"normal","swap_used_bytes":6997540864},"model":"Llama-3.2-3B-Instruct-q4f16_1-MLC","prefill":[{"completion_tokens":23,"decode_tps":18.8,"e2e_s":14.839,"prefill_tps":107.1,"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":13.614,"wall_s":14.84}],"seq":28,"t":80344,"type":"model_speed","was_cached":true}}
nightshift: [1792084302] page: {"bench":{"items":0,"model":"Llama-3.2-3B-Instruct-q4f16_1-MLC","seconds":0,"seq":29,"t":80349,"type":"suite_done"}}
nightshift: [1792084315] page: {"bench":{"label":"load Phi-3.5-mini-instruct-q4f16_1-MLC","seq":30,"t":93591,"text":"Start to fetch params","type":"progress"}}
nightshift: [1792084316] page: {"bench":{"label":"load Phi-3.5-mini-instruct-q4f16_1-MLC","seq":31,"t":94254,"text":"Loading model from cache[8/83]: 215MB loaded. 10% completed, 14 secs elapsed.","type":"progress"}}
nightshift: [1792084316] page: {"bench":{"label":"load Phi-3.5-mini-instruct-q4f16_1-MLC","seq":32,"t":94845,"text":"Loading model from cache[16/83]: 413MB loaded. 20% completed, 14 secs elapsed.","type":"progress"}}
nightshift: [1792084317] page: {"bench":{"label":"load Phi-3.5-mini-instruct-q4f16_1-MLC","seq":33,"t":95589,"text":"Loading model from cache[25/83]: 636MB loaded. 30% completed, 15 secs elapsed.","type":"progress"}}
nightshift: [1792084318] page: {"bench":{"label":"load Phi-3.5-mini-instruct-q4f16_1-MLC","seq":34,"t":96258,"text":"Loading model from cache[33/83]: 839MB loaded. 40% completed, 16 secs elapsed.","type":"progress"}}
nightshift: [1792084318] page: {"bench":{"label":"load Phi-3.5-mini-instruct-q4f16_1-MLC","seq":35,"t":96921,"text":"Loading model from cache[41/83]: 1037MB loaded. 50% completed, 16 secs elapsed.","type":"progress"}}
nightshift: [1792084319] page: {"bench":{"label":"load Phi-3.5-mini-instruct-q4f16_1-MLC","seq":36,"t":97578,"text":"Loading model from cache[49/83]: 1235MB loaded. 60% completed, 17 secs elapsed.","type":"progress"}}
nightshift: [1792084320] page: {"bench":{"label":"load Phi-3.5-mini-instruct-q4f16_1-MLC","seq":37,"t":98238,"text":"Loading model from cache[58/83]: 1447MB loaded. 70% completed, 18 secs elapsed.","type":"progress"}}
nightshift: [1792084320] page: {"bench":{"label":"load Phi-3.5-mini-instruct-q4f16_1-MLC","seq":38,"t":98809,"text":"Loading model from cache[66/83]: 1649MB loaded. 80% completed, 18 secs elapsed.","type":"progress"}}
nightshift: [1792084321] page: {"bench":{"label":"load Phi-3.5-mini-instruct-q4f16_1-MLC","seq":39,"t":99273,"text":"Loading model from cache[74/83]: 1848MB loaded. 90% completed, 19 secs elapsed.","type":"progress"}}
nightshift: [1792084321] page: {"bench":{"label":"load Phi-3.5-mini-instruct-q4f16_1-MLC","seq":40,"t":99853,"text":"Loading model from cache[83/83]: 2051MB loaded. 100% completed, 19 secs elapsed.","type":"progress"}}
nightshift: [1792084344] page: {"bench":{"decode":[{"completion_tokens":15,"decode_tps":18.5,"e2e_s":1.593,"prefill_tps":80,"prompt_tokens":62,"text":"The bicycle, a marvel of human ingenuity and a symbol","ttft_s":0.775,"wall_s":1.594}],"first_call":{"ttft_s":0.528,"wall_s":0.787},"load_s":18.989,"mem_after_load":{"free_pct":48,"pressure":"normal","swap_used_bytes":7363821568},"mem_after_speed":{"free_pct":44,"pressure":"normal","swap_used_bytes":7263158272},"mem_before":{"free_pct":48,"pressure":"normal","swap_used_bytes":7607222272},"model":"Phi-3.5-mini-instruct-q4f16_1-MLC","prefill":[{"completion_tokens":23,"decode_tps":16.5,"e2e_s":20.098,"prefill_tps":86.8,"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":18.7,"wall_s":20.099}],"seq":41,"t":122360,"type":"model_speed","was_cached":true}}
nightshift: [1792084344] page: {"bench":{"items":0,"model":"Phi-3.5-mini-instruct-q4f16_1-MLC","seconds":0,"seq":42,"t":122378,"type":"suite_done"}}
nightshift: [1792084360] page: {"bench":{"label":"load SmolLM2-1.7B-Instruct-q4f16_1-MLC","seq":43,"t":138231,"text":"Start to fetch params","type":"progress"}}
nightshift: [1792084360] page: {"bench":{"label":"load SmolLM2-1.7B-Instruct-q4f16_1-MLC","seq":44,"t":138570,"text":"Loading model from cache[3/37]: 96MB loaded. 10% completed, 16 secs elapsed.","type":"progress"}}
nightshift: [1792084360] page: {"bench":{"label":"load SmolLM2-1.7B-Instruct-q4f16_1-MLC","seq":45,"t":138880,"text":"Loading model from cache[7/37]: 199MB loaded. 21% completed, 16 secs elapsed.","type":"progress"}}
nightshift: [1792084361] page: {"bench":{"label":"load SmolLM2-1.7B-Instruct-q4f16_1-MLC","seq":46,"t":139221,"text":"Loading model from cache[11/37]: 298MB loaded. 32% completed, 17 secs elapsed.","type":"progress"}}
nightshift: [1792084361] page: {"bench":{"label":"load SmolLM2-1.7B-Instruct-q4f16_1-MLC","seq":47,"t":139615,"text":"Loading model from cache[14/37]: 370MB loaded. 40% completed, 17 secs elapsed.","type":"progress"}}
nightshift: [1792084361] page: {"bench":{"label":"load SmolLM2-1.7B-Instruct-q4f16_1-MLC","seq":48,"t":139893,"text":"Loading model from cache[19/37]: 487MB loaded. 52% completed, 17 secs elapsed.","type":"progress"}}
nightshift: [1792084362] page: {"bench":{"label":"load SmolLM2-1.7B-Instruct-q4f16_1-MLC","seq":49,"t":140074,"text":"Loading model from cache[22/37]: 559MB loaded. 60% completed, 17 secs elapsed.","type":"progress"}}
nightshift: [1792084362] page: {"bench":{"label":"load SmolLM2-1.7B-Instruct-q4f16_1-MLC","seq":50,"t":140363,"text":"Loading model from cache[26/37]: 658MB loaded. 71% completed, 18 secs elapsed.","type":"progress"}}
nightshift: [1792084362] page: {"bench":{"label":"load SmolLM2-1.7B-Instruct-q4f16_1-MLC","seq":51,"t":140587,"text":"Loading model from cache[30/37]: 746MB loaded. 81% completed, 18 secs elapsed.","type":"progress"}}
nightshift: [1792084362] page: {"bench":{"label":"load SmolLM2-1.7B-Instruct-q4f16_1-MLC","seq":52,"t":140813,"text":"Loading model from cache[34/37]: 847MB loaded. 92% completed, 18 secs elapsed.","type":"progress"}}
nightshift: [1792084363] page: {"bench":{"label":"load SmolLM2-1.7B-Instruct-q4f16_1-MLC","seq":53,"t":141022,"text":"Loading model from cache[37/37]: 919MB loaded. 100% completed, 18 secs elapsed.","type":"progress"}}
nightshift: [1792084372] page: {"bench":{"decode":[{"completion_tokens":15,"decode_tps":33,"e2e_s":0.752,"prefill_tps":221.5,"prompt_tokens":64,"text":"The bicycle, a mode of transportation that has evolved significantly over the centuries, has","ttft_s":0.289,"wall_s":0.767}],"first_call":{"ttft_s":0.383,"wall_s":0.54},"load_s":18.147,"mem_after_load":{"free_pct":59,"pressure":"normal","swap_used_bytes":7154106368},"mem_after_speed":{"free_pct":51,"pressure":"normal","swap_used_bytes":7112163328},"mem_before":{"free_pct":44,"pressure":"normal","swap_used_bytes":7263158272},"model":"SmolLM2-1.7B-Instruct-q4f16_1-MLC","prefill":[{"completion_tokens":23,"decode_tps":35.5,"e2e_s":8.419,"prefill_tps":195.9,"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":7.764,"wall_s":8.429}],"seq":54,"t":150793,"type":"model_speed","was_cached":true}}
nightshift: [1792084372] page: {"bench":{"items":0,"model":"SmolLM2-1.7B-Instruct-q4f16_1-MLC","seconds":0,"seq":55,"t":150807,"type":"suite_done"}}
nightshift: [1792084390] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":56,"t":167976,"text":"Start to fetch params","type":"progress"}}
nightshift: [1792084391] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":57,"t":168939,"text":"Loading model from cache[1/42]: 282MB loaded. 20% completed, 17 secs elapsed.","type":"progress"}}
nightshift: [1792084391] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":58,"t":169272,"text":"Loading model from cache[6/42]: 432MB loaded. 30% completed, 17 secs elapsed.","type":"progress"}}
nightshift: [1792084391] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":59,"t":169575,"text":"Loading model from cache[11/42]: 568MB loaded. 40% completed, 17 secs elapsed.","type":"progress"}}
nightshift: [1792084391] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":60,"t":169898,"text":"Loading model from cache[16/42]: 703MB loaded. 50% completed, 18 secs elapsed.","type":"progress"}}
nightshift: [1792084392] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":61,"t":170275,"text":"Loading model from cache[21/42]: 850MB loaded. 60% completed, 18 secs elapsed.","type":"progress"}}
nightshift: [1792084392] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":62,"t":170568,"text":"Loading model from cache[26/42]: 985MB loaded. 70% completed, 18 secs elapsed.","type":"progress"}}
nightshift: [1792084392] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":63,"t":170918,"text":"Loading model from cache[31/42]: 1131MB loaded. 80% completed, 19 secs elapsed.","type":"progress"}}
nightshift: [1792084393] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":64,"t":171241,"text":"Loading model from cache[36/42]: 1267MB loaded. 90% completed, 19 secs elapsed.","type":"progress"}}
nightshift: [1792084393] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":65,"t":171624,"text":"Loading model from cache[42/42]: 1403MB loaded. 100% completed, 19 secs elapsed.","type":"progress"}}
nightshift: [1792084407] page: {"bench":{"decode":[{"completion_tokens":15,"decode_tps":23.9,"e2e_s":0.974,"prefill_tps":122.4,"prompt_tokens":42,"text":"## The Pedal Revolution: A Century of Cycling\n\nThe bicycle, a seemingly simple","ttft_s":0.343,"wall_s":0.974}],"first_call":{"ttft_s":0.45,"wall_s":0.636},"load_s":20.288,"mem_after_load":{"free_pct":57,"pressure":"normal","swap_used_bytes":6810173440},"mem_after_speed":{"free_pct":45,"pressure":"normal","swap_used_bytes":6793396224},"mem_before":{"free_pct":54,"pressure":"normal","swap_used_bytes":7086997504},"model":"gemma-2-2b-it-q4f16_1-MLC","prefill":[{"completion_tokens":23,"decode_tps":20.7,"e2e_s":12.215,"prefill_tps":134.2,"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":11.1,"wall_s":12.219}],"seq":66,"t":185473,"type":"model_speed","was_cached":true}}
nightshift: [1792084407] page: {"bench":{"items":0,"model":"gemma-2-2b-it-q4f16_1-MLC","seconds":0,"seq":67,"t":185477,"type":"suite_done"}}
nightshift: [1792084422] page: {"bench":{"label":"load Qwen2.5-Coder-3B-Instruct-q4f16_1-MLC","seq":68,"t":200528,"text":"Start to fetch params","type":"progress"}}
nightshift: [1792084423] page: {"bench":{"label":"load Qwen2.5-Coder-3B-Instruct-q4f16_1-MLC","seq":69,"t":201276,"text":"Loading model from cache[2/62]: 170MB loaded. 10% completed, 15 secs elapsed.","type":"progress"}}
nightshift: [1792084423] page: {"bench":{"label":"load Qwen2.5-Coder-3B-Instruct-q4f16_1-MLC","seq":70,"t":201798,"text":"Loading model from cache[9/62]: 347MB loaded. 20% completed, 16 secs elapsed.","type":"progress"}}
nightshift: [1792084424] page: {"bench":{"label":"load Qwen2.5-Coder-3B-Instruct-q4f16_1-MLC","seq":71,"t":202264,"text":"Loading model from cache[15/62]: 498MB loaded. 30% completed, 16 secs elapsed.","type":"progress"}}
nightshift: [1792084424] page: {"bench":{"label":"load Qwen2.5-Coder-3B-Instruct-q4f16_1-MLC","seq":72,"t":202712,"text":"Loading model from cache[22/62]: 665MB loaded. 40% completed, 17 secs elapsed.","type":"progress"}}
nightshift: [1792084425] page: {"bench":{"label":"load Qwen2.5-Coder-3B-Instruct-q4f16_1-MLC","seq":73,"t":203205,"text":"Loading model from cache[29/62]: 843MB loaded. 50% completed, 17 secs elapsed.","type":"progress"}}
nightshift: [1792084425] page: {"bench":{"label":"load Qwen2.5-Coder-3B-Instruct-q4f16_1-MLC","seq":74,"t":203652,"text":"Loading model from cache[35/62]: 995MB loaded. 60% completed, 18 secs elapsed.","type":"progress"}}
nightshift: [1792084426] page: {"bench":{"label":"load Qwen2.5-Coder-3B-Instruct-q4f16_1-MLC","seq":75,"t":204124,"text":"Loading model from cache[42/62]: 1165MB loaded. 70% completed, 18 secs elapsed.","type":"progress"}}
nightshift: [1792084426] page: {"bench":{"label":"load Qwen2.5-Coder-3B-Instruct-q4f16_1-MLC","seq":76,"t":204589,"text":"Loading model from cache[49/62]: 1332MB loaded. 80% completed, 19 secs elapsed.","type":"progress"}}
nightshift: [1792084427] page: {"bench":{"label":"load Qwen2.5-Coder-3B-Instruct-q4f16_1-MLC","seq":77,"t":205110,"text":"Loading model from cache[56/62]: 1505MB loaded. 90% completed, 19 secs elapsed.","type":"progress"}}
nightshift: [1792084427] page: {"bench":{"label":"load Qwen2.5-Coder-3B-Instruct-q4f16_1-MLC","seq":78,"t":205593,"text":"Loading model from cache[62/62]: 1656MB loaded. 100% completed, 20 secs elapsed.","type":"progress"}}
nightshift: [1792084447] page: {"bench":{"decode":[{"completion_tokens":15,"decode_tps":17.7,"e2e_s":1.393,"prefill_tps":97.4,"prompt_tokens":53,"text":"The bicycle has a rich and fascinating history that spans over two centuries. The first","ttft_s":0.544,"wall_s":1.398}],"first_call":{"ttft_s":0.63,"wall_s":0.929},"load_s":19.607,"mem_after_load":{"free_pct":46,"pressure":"normal","swap_used_bytes":5835194368},"mem_after_speed":{"free_pct":47,"pressure":"normal","swap_used_bytes":5514264576},"mem_before":{"free_pct":46,"pressure":"normal","swap_used_bytes":6793396224},"model":"Qwen2.5-Coder-3B-Instruct-q4f16_1-MLC","prefill":[{"completion_tokens":23,"decode_tps":13.8,"e2e_s":17.248,"prefill_tps":95.5,"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":15.576,"wall_s":17.249}],"seq":79,"t":225208,"type":"model_speed","was_cached":true}}
nightshift: [1792084447] page: {"bench":{"items":0,"model":"Qwen2.5-Coder-3B-Instruct-q4f16_1-MLC","seconds":0,"seq":80,"t":225211,"type":"suite_done"}}
nightshift: [1792084447] page: {"bench":{"phase":"models","seconds":222.24,"seq":81,"t":225214,"type":"phase_end"}}
nightshift: [1792084447] page: {"bench":{"errors":[],"seconds":222.338,"seq":82,"t":225216,"type":"done"}}