nightshift: [1792084457] start v0.90.0 · Apple M5 (m5+) · 16 GB · macOS 26.4 · 4P+6E cores
nightshift: [1792084457] serving /Users/samuelcummings/nightshift-m1m4-build/web-bench on http://127.0.0.1:47321/
nightshift: [1792084457] webkit: BackgroundWebContentRunningBoardThrottlingEnabled=false, HiddenPageDOMTimerThrottlingAutoIncreases=false, HiddenPageDOMTimerThrottlingEnabled=false, PageVisibilityBasedProcessSuppressionEnabled=false, WebGPUEnabled=true
nightshift: [1792084457] webkit: window occlusion detection off (page stays active behind other windows): true
nightshift: [1792084457] cockpit loading (injected scripts removed: true)
nightshift: [1792084457] quit gate installed (applicationShouldTerminate:)
nightshift: [1792084460] page: {"bench":{"label":"97-loadmem","phases":["env","models"],"seq":0,"t":2861,"type":"start"}}
nightshift: [1792084460] page: {"bench":{"phase":"env","seq":1,"t":2870,"type":"phase_start"}}
nightshift: [1792084460] page: {"bench":{"boot":{"bench_script_ms":155,"dom_content_loaded_ms":304,"engine_ready_ms":224,"load_event_ms":356,"resources":211,"scripts":175,"shell_mounted_ms":362,"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":68,"pressure":"normal","swap_used_bytes":5430378496},"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":3012,"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":0.127}}}
nightshift: [1792084460] page: {"bench":{"phase":"env","seconds":0.146,"seq":3,"t":3016,"type":"phase_end"}}
nightshift: [1792084460] page: {"bench":{"phase":"models","seq":4,"t":3018,"type":"phase_start"}}
nightshift: [1792084466] 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: [1792084477] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":5,"t":19548,"text":"Start to fetch params","type":"progress"}}
nightshift: [1792084478] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":6,"t":20679,"text":"Loading model from cache[1/42]: 282MB loaded. 20% completed, 16 secs elapsed.","type":"progress"}}
nightshift: [1792084478] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":7,"t":21093,"text":"Loading model from cache[6/42]: 432MB loaded. 30% completed, 16 secs elapsed.","type":"progress"}}
nightshift: [1792084479] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":8,"t":21624,"text":"Loading model from cache[11/42]: 568MB loaded. 40% completed, 17 secs elapsed.","type":"progress"}}
nightshift: [1792084479] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":9,"t":22193,"text":"Loading model from cache[16/42]: 703MB loaded. 50% completed, 18 secs elapsed.","type":"progress"}}
nightshift: [1792084480] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":10,"t":22689,"text":"Loading model from cache[21/42]: 850MB loaded. 60% completed, 18 secs elapsed.","type":"progress"}}
nightshift: [1792084480] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":11,"t":23061,"text":"Loading model from cache[26/42]: 985MB loaded. 70% completed, 18 secs elapsed.","type":"progress"}}
nightshift: [1792084481] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":12,"t":23366,"text":"Loading model from cache[31/42]: 1131MB loaded. 80% completed, 19 secs elapsed.","type":"progress"}}
nightshift: [1792084481] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":13,"t":23695,"text":"Loading model from cache[36/42]: 1267MB loaded. 90% completed, 19 secs elapsed.","type":"progress"}}
nightshift: [1792084481] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":14,"t":23992,"text":"Loading model from cache[42/42]: 1403MB loaded. 100% completed, 19 secs elapsed.","type":"progress"}}
nightshift: [1792084496] page: {"bench":{"decode":[{"completion_tokens":15,"decode_tps":20,"e2e_s":1.14,"prefill_tps":109.7,"prompt_tokens":42,"text":"## The Pedal Revolution: A Century of Cycling\n\nThe bicycle, a seemingly simple","ttft_s":0.383,"wall_s":1.14}],"first_call":{"ttft_s":0.385,"wall_s":0.619},"load_s":20.473,"mem_after_load":{"free_pct":42,"pressure":"warn","swap_used_bytes":5505417216},"mem_after_speed":{"free_pct":34,"pressure":"warn","swap_used_bytes":5488640000},"mem_before":{"free_pct":67,"pressure":"normal","swap_used_bytes":5421989888},"model":"gemma-2-2b-it-q4f16_1-MLC","prefill":[{"completion_tokens":23,"decode_tps":18,"e2e_s":13.14,"prefill_tps":125.7,"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":11.849,"wall_s":13.145}],"seq":15,"t":38922,"type":"model_speed","was_cached":true}}
nightshift: [1792084496] page: {"bench":{"items":0,"model":"gemma-2-2b-it-q4f16_1-MLC","seconds":0,"seq":16,"t":38940,"type":"suite_done"}}
nightshift: [1792084513] page: {"bench":{"label":"load Llama-3.2-3B-Instruct-q4f16_1-MLC","seq":17,"t":55817,"text":"Start to fetch params","type":"progress"}}
nightshift: [1792084514] page: {"bench":{"label":"load Llama-3.2-3B-Instruct-q4f16_1-MLC","seq":18,"t":56432,"text":"Loading model from cache[1/58]: 188MB loaded. 10% completed, 17 secs elapsed.","type":"progress"}}
nightshift: [1792084514] page: {"bench":{"label":"load Llama-3.2-3B-Instruct-q4f16_1-MLC","seq":19,"t":56950,"text":"Loading model from cache[8/58]: 374MB loaded. 21% completed, 17 secs elapsed.","type":"progress"}}
nightshift: [1792084515] page: {"bench":{"label":"load Llama-3.2-3B-Instruct-q4f16_1-MLC","seq":20,"t":57389,"text":"Loading model from cache[14/58]: 536MB loaded. 31% completed, 18 secs elapsed.","type":"progress"}}
nightshift: [1792084515] page: {"bench":{"label":"load Llama-3.2-3B-Instruct-q4f16_1-MLC","seq":21,"t":57761,"text":"Loading model from cache[20/58]: 698MB loaded. 40% completed, 18 secs elapsed.","type":"progress"}}
nightshift: [1792084516] page: {"bench":{"label":"load Llama-3.2-3B-Instruct-q4f16_1-MLC","seq":22,"t":58244,"text":"Loading model from cache[27/58]: 884MB loaded. 51% completed, 19 secs elapsed.","type":"progress"}}
nightshift: [1792084516] page: {"bench":{"label":"load Llama-3.2-3B-Instruct-q4f16_1-MLC","seq":23,"t":58657,"text":"Loading model from cache[33/58]: 1046MB loaded. 60% completed, 19 secs elapsed.","type":"progress"}}
nightshift: [1792084516] page: {"bench":{"label":"load Llama-3.2-3B-Instruct-q4f16_1-MLC","seq":24,"t":59134,"text":"Loading model from cache[39/58]: 1208MB loaded. 70% completed, 20 secs elapsed.","type":"progress"}}
nightshift: [1792084517] page: {"bench":{"label":"load Llama-3.2-3B-Instruct-q4f16_1-MLC","seq":25,"t":59683,"text":"Loading model from cache[46/58]: 1400MB loaded. 81% completed, 20 secs elapsed.","type":"progress"}}
nightshift: [1792084517] page: {"bench":{"label":"load Llama-3.2-3B-Instruct-q4f16_1-MLC","seq":26,"t":60120,"text":"Loading model from cache[52/58]: 1562MB loaded. 90% completed, 21 secs elapsed.","type":"progress"}}
nightshift: [1792084518] page: {"bench":{"label":"load Llama-3.2-3B-Instruct-q4f16_1-MLC","seq":27,"t":60709,"text":"Loading model from cache[58/58]: 1724MB loaded. 100% completed, 21 secs elapsed.","type":"progress"}}
nightshift: [1792084536] page: {"bench":{"decode":[{"completion_tokens":15,"decode_tps":18.7,"e2e_s":1.413,"prefill_tps":89.1,"prompt_tokens":54,"text":"The bicycle, a ubiquitous mode of transportation and recreation, has a rich and fascinating","ttft_s":0.606,"wall_s":1.417}],"first_call":{"ttft_s":0.545,"wall_s":0.821},"load_s":21.216,"mem_after_load":{"free_pct":36,"pressure":"warn","swap_used_bytes":5429919744},"mem_after_speed":{"free_pct":26,"pressure":"warn","swap_used_bytes":5413142528},"mem_before":{"free_pct":34,"pressure":"warn","swap_used_bytes":5488640000},"model":"Llama-3.2-3B-Instruct-q4f16_1-MLC","prefill":[{"completion_tokens":23,"decode_tps":18.2,"e2e_s":15.811,"prefill_tps":100.3,"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":14.536,"wall_s":15.811}],"seq":28,"t":78799,"type":"model_speed","was_cached":true}}
nightshift: [1792084536] page: {"bench":{"items":0,"model":"Llama-3.2-3B-Instruct-q4f16_1-MLC","seconds":0,"seq":29,"t":78815,"type":"suite_done"}}
nightshift: [1792084537] page: {"trim":{"at":1792084537131,"budgetMb":3600,"dropped":["Phi-3.5-mini-instruct-q4f16_1-MLC"],"keep":[],"maxModels":1,"panelMb":3000,"pressure":"warn","refused":true,"usedMb":0}}
nightshift: [1792084537] page: {"bench":{"error":"memory_pressure memory_pressure: this Mac is under warn memory pressure — device lanes wait; close other apps or use a cloud lane","mem_before":{"free_pct":26,"pressure":"warn","swap_used_bytes":5404753920},"model":"Phi-3.5-mini-instruct-q4f16_1-MLC","seq":30,"t":79360,"type":"model_speed","was_cached":true}}
nightshift: [1792084553] page: {"bench":{"label":"load SmolLM2-1.7B-Instruct-q4f16_1-MLC","seq":31,"t":95595,"text":"Start to fetch params","type":"progress"}}
nightshift: [1792084553] page: {"bench":{"label":"load SmolLM2-1.7B-Instruct-q4f16_1-MLC","seq":32,"t":95853,"text":"Loading model from cache[3/37]: 96MB loaded. 10% completed, 16 secs elapsed.","type":"progress"}}
nightshift: [1792084553] page: {"bench":{"label":"load SmolLM2-1.7B-Instruct-q4f16_1-MLC","seq":33,"t":96211,"text":"Loading model from cache[7/37]: 199MB loaded. 21% completed, 17 secs elapsed.","type":"progress"}}
nightshift: [1792084554] page: {"bench":{"label":"load SmolLM2-1.7B-Instruct-q4f16_1-MLC","seq":34,"t":96797,"text":"Loading model from cache[11/37]: 298MB loaded. 32% completed, 17 secs elapsed.","type":"progress"}}
nightshift: [1792084554] page: {"bench":{"label":"load SmolLM2-1.7B-Instruct-q4f16_1-MLC","seq":35,"t":97054,"text":"Loading model from cache[14/37]: 370MB loaded. 40% completed, 17 secs elapsed.","type":"progress"}}
nightshift: [1792084555] page: {"bench":{"label":"load SmolLM2-1.7B-Instruct-q4f16_1-MLC","seq":36,"t":97470,"text":"Loading model from cache[19/37]: 487MB loaded. 52% completed, 18 secs elapsed.","type":"progress"}}
nightshift: [1792084555] page: {"bench":{"label":"load SmolLM2-1.7B-Instruct-q4f16_1-MLC","seq":37,"t":97708,"text":"Loading model from cache[22/37]: 559MB loaded. 60% completed, 18 secs elapsed.","type":"progress"}}
nightshift: [1792084555] page: {"bench":{"label":"load SmolLM2-1.7B-Instruct-q4f16_1-MLC","seq":38,"t":98054,"text":"Loading model from cache[26/37]: 658MB loaded. 71% completed, 18 secs elapsed.","type":"progress"}}
nightshift: [1792084556] page: {"bench":{"label":"load SmolLM2-1.7B-Instruct-q4f16_1-MLC","seq":39,"t":98422,"text":"Loading model from cache[30/37]: 746MB loaded. 81% completed, 19 secs elapsed.","type":"progress"}}
nightshift: [1792084556] page: {"bench":{"label":"load SmolLM2-1.7B-Instruct-q4f16_1-MLC","seq":40,"t":98788,"text":"Loading model from cache[34/37]: 847MB loaded. 92% completed, 19 secs elapsed.","type":"progress"}}
nightshift: [1792084556] page: {"bench":{"label":"load SmolLM2-1.7B-Instruct-q4f16_1-MLC","seq":41,"t":99075,"text":"Loading model from cache[37/37]: 919MB loaded. 100% completed, 20 secs elapsed.","type":"progress"}}
nightshift: [1792084566] page: {"bench":{"decode":[{"completion_tokens":15,"decode_tps":38.3,"e2e_s":0.68,"prefill_tps":227,"prompt_tokens":64,"text":"The bicycle, a mode of transportation that has evolved significantly over the centuries, has","ttft_s":0.282,"wall_s":0.68}],"first_call":{"ttft_s":0.438,"wall_s":0.595},"load_s":19.231,"mem_after_load":{"free_pct":57,"pressure":"normal","swap_used_bytes":6447104000},"mem_after_speed":{"free_pct":51,"pressure":"normal","swap_used_bytes":6371606528},"mem_before":{"free_pct":25,"pressure":"warn","swap_used_bytes":5404753920},"model":"SmolLM2-1.7B-Instruct-q4f16_1-MLC","prefill":[{"completion_tokens":23,"decode_tps":36.2,"e2e_s":8.767,"prefill_tps":187.2,"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":8.126,"wall_s":8.773}],"seq":42,"t":109159,"type":"model_speed","was_cached":true}}
nightshift: [1792084566] page: {"bench":{"items":0,"model":"SmolLM2-1.7B-Instruct-q4f16_1-MLC","seconds":0,"seq":43,"t":109171,"type":"suite_done"}}
nightshift: [1792084581] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":44,"t":123732,"text":"Start to fetch params","type":"progress"}}
nightshift: [1792084582] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":45,"t":124823,"text":"Loading model from cache[1/42]: 282MB loaded. 20% completed, 15 secs elapsed.","type":"progress"}}
nightshift: [1792084582] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":46,"t":125192,"text":"Loading model from cache[6/42]: 432MB loaded. 30% completed, 15 secs elapsed.","type":"progress"}}
nightshift: [1792084583] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":47,"t":125521,"text":"Loading model from cache[11/42]: 568MB loaded. 40% completed, 15 secs elapsed.","type":"progress"}}
nightshift: [1792084583] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":48,"t":125875,"text":"Loading model from cache[16/42]: 703MB loaded. 50% completed, 16 secs elapsed.","type":"progress"}}
nightshift: [1792084583] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":49,"t":126204,"text":"Loading model from cache[21/42]: 850MB loaded. 60% completed, 16 secs elapsed.","type":"progress"}}
nightshift: [1792084584] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":50,"t":126500,"text":"Loading model from cache[26/42]: 985MB loaded. 70% completed, 16 secs elapsed.","type":"progress"}}
nightshift: [1792084584] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":51,"t":126884,"text":"Loading model from cache[31/42]: 1131MB loaded. 80% completed, 17 secs elapsed.","type":"progress"}}
nightshift: [1792084584] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":52,"t":127217,"text":"Loading model from cache[36/42]: 1267MB loaded. 90% completed, 17 secs elapsed.","type":"progress"}}
nightshift: [1792084585] page: {"bench":{"label":"load gemma-2-2b-it-q4f16_1-MLC","seq":53,"t":127551,"text":"Loading model from cache[42/42]: 1403MB loaded. 100% completed, 17 secs elapsed.","type":"progress"}}
nightshift: [1792084599] page: {"bench":{"decode":[{"completion_tokens":15,"decode_tps":20.4,"e2e_s":1.095,"prefill_tps":117.6,"prompt_tokens":42,"text":"## The Pedal Revolution: A Century of Cycling\n\nThe bicycle, a seemingly simple","ttft_s":0.357,"wall_s":1.095}],"first_call":{"ttft_s":0.448,"wall_s":0.628},"load_s":17.873,"mem_after_load":{"free_pct":57,"pressure":"normal","swap_used_bytes":6262554624},"mem_after_speed":{"free_pct":44,"pressure":"normal","swap_used_bytes":6254166016},"mem_before":{"free_pct":52,"pressure":"normal","swap_used_bytes":6371606528},"model":"gemma-2-2b-it-q4f16_1-MLC","prefill":[{"completion_tokens":23,"decode_tps":21.3,"e2e_s":12.8,"prefill_tps":127.2,"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":11.714,"wall_s":12.799}],"seq":54,"t":142093,"type":"model_speed","was_cached":true}}
nightshift: [1792084599] page: {"bench":{"items":0,"model":"gemma-2-2b-it-q4f16_1-MLC","seconds":0,"seq":55,"t":142100,"type":"suite_done"}}
nightshift: [1792084613] page: {"bench":{"label":"load Qwen2.5-Coder-3B-Instruct-q4f16_1-MLC","seq":56,"t":155620,"text":"Start to fetch params","type":"progress"}}
nightshift: [1792084613] page: {"bench":{"label":"load Qwen2.5-Coder-3B-Instruct-q4f16_1-MLC","seq":57,"t":156133,"text":"Loading model from cache[2/62]: 170MB loaded. 10% completed, 13 secs elapsed.","type":"progress"}}
nightshift: [1792084614] page: {"bench":{"label":"load Qwen2.5-Coder-3B-Instruct-q4f16_1-MLC","seq":58,"t":156674,"text":"Loading model from cache[9/62]: 347MB loaded. 20% completed, 14 secs elapsed.","type":"progress"}}
nightshift: [1792084615] page: {"bench":{"label":"load Qwen2.5-Coder-3B-Instruct-q4f16_1-MLC","seq":59,"t":157341,"text":"Loading model from cache[15/62]: 498MB loaded. 30% completed, 15 secs elapsed.","type":"progress"}}
nightshift: [1792084615] page: {"bench":{"label":"load Qwen2.5-Coder-3B-Instruct-q4f16_1-MLC","seq":60,"t":158219,"text":"Loading model from cache[22/62]: 665MB loaded. 40% completed, 16 secs elapsed.","type":"progress"}}
nightshift: [1792084617] page: {"bench":{"label":"load Qwen2.5-Coder-3B-Instruct-q4f16_1-MLC","seq":61,"t":159603,"text":"Loading model from cache[29/62]: 843MB loaded. 50% completed, 17 secs elapsed.","type":"progress"}}
nightshift: [1792084618] page: {"bench":{"label":"load Qwen2.5-Coder-3B-Instruct-q4f16_1-MLC","seq":62,"t":160335,"text":"Loading model from cache[35/62]: 995MB loaded. 60% completed, 18 secs elapsed.","type":"progress"}}
nightshift: [1792084618] page: {"bench":{"label":"load Qwen2.5-Coder-3B-Instruct-q4f16_1-MLC","seq":63,"t":160800,"text":"Loading model from cache[42/62]: 1165MB loaded. 70% completed, 18 secs elapsed.","type":"progress"}}
nightshift: [1792084619] page: {"bench":{"label":"load Qwen2.5-Coder-3B-Instruct-q4f16_1-MLC","seq":64,"t":161288,"text":"Loading model from cache[49/62]: 1332MB loaded. 80% completed, 19 secs elapsed.","type":"progress"}}
nightshift: [1792084619] page: {"bench":{"label":"load Qwen2.5-Coder-3B-Instruct-q4f16_1-MLC","seq":65,"t":161729,"text":"Loading model from cache[56/62]: 1505MB loaded. 90% completed, 19 secs elapsed.","type":"progress"}}
nightshift: [1792084619] page: {"bench":{"label":"load Qwen2.5-Coder-3B-Instruct-q4f16_1-MLC","seq":66,"t":162102,"text":"Loading model from cache[62/62]: 1656MB loaded. 100% completed, 19 secs elapsed.","type":"progress"}}
nightshift: [1792084637] page: {"bench":{"decode":[{"completion_tokens":15,"decode_tps":18.7,"e2e_s":1.475,"prefill_tps":79.7,"prompt_tokens":53,"text":"The bicycle has a rich and fascinating history that spans over two centuries. The first","ttft_s":0.665,"wall_s":1.478}],"first_call":{"ttft_s":0.425,"wall_s":0.702},"load_s":19.481,"mem_after_load":{"free_pct":50,"pressure":"normal","swap_used_bytes":6136725504},"mem_after_speed":{"free_pct":45,"pressure":"normal","swap_used_bytes":6103171072},"mem_before":{"free_pct":44,"pressure":"normal","swap_used_bytes":6254166016},"model":"Qwen2.5-Coder-3B-Instruct-q4f16_1-MLC","prefill":[{"completion_tokens":23,"decode_tps":13.3,"e2e_s":15.421,"prefill_tps":108.6,"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":13.69,"wall_s":15.425}],"seq":67,"t":179729,"type":"model_speed","was_cached":true}}
nightshift: [1792084637] page: {"bench":{"items":0,"model":"Qwen2.5-Coder-3B-Instruct-q4f16_1-MLC","seconds":0,"seq":68,"t":179734,"type":"suite_done"}}
nightshift: [1792084637] page: {"bench":{"phase":"models","seconds":176.727,"seq":69,"t":179745,"type":"phase_end"}}
nightshift: [1792084637] page: {"bench":{"errors":[],"seconds":176.878,"seq":70,"t":179748,"type":"done"}}