{ "amendments": [ { "id": "A-L01", "new": "swap used < 80 % AND MemAvailable >= 10 GiB AND PSI memory full avg60 < 10 % (/dev/shm < 4 GiB kept)", "old": "swap used < 50 %, /dev/shm used < 4 GiB, MemAvailable >= 10 GiB", "readings_at_1824Z": { "MemAvailable_GiB": 49.2, "MemFree_GiB": 41.0, "bench_unit_MemorySwapCurrent_bytes": 2994176, "largest_swap_holders": "two stale sharpsignal.memory_mcp processes (pids 9106, 12122; ~1.1 GB each) in old ghostty scopes, and the system ollama's nomic runner (0.34 GB)", "psi_memory_full_avg60_pct": 1.74, "swap_img": "0.3 of 2.0 GiB", "zram0": "9.9 of 16.0 GiB used, stored in 3.56 GiB of RAM" }, "reason": "the old gate tripped at 18:24:03Z (dry12-anchor-4096/REFUSED.json: swap 54.2 %) with no memory risk: this box's swap is mostly zram (compressed RAM), so a bare percentage over-reads the risk; the new gate still catches the 15:56Z case (swap 17 of 17 GiB)", "rows_the_author_had_seen": "none scored: the gate tripped before item 12's first load", "ruled_by": "the lead under R-0 (message to the laptop lane, received ~18:25Z; applied 18:25:59Z)", "utc": "2026-09-23T18:25Z", "what": "the memory gate (PLAN-v2 §7, G-2)" }, { "id": "A-L02", "new": "MemoryMax=16G", "old": "MemoryMax=4G", "reason": "item 14 measured: under 4G the serve unit hit its cap 633,419 times (memory.events max) during item 12, swapped up to 6.0 GB (MemorySwapPeak 6,008,573,952 B), memory.pressure full avg60 read 9.41 %, and prefill stalled 11,416 ms on doorman item A5 and 4,607 ms on C1 (515-521 prompt tokens; the other items 121-342 ms). No OOM, but the cap throttled the measurement. 16G is a ceiling that does not bind: llama-server's host prompt cache is bounded (~8 GiB) and the rest is clean page cache; each arm.json records memory.events, memory.stat and memory.pressure before and after, so a binding cap is visible", "rows_the_author_had_seen": "dry12-anchor-4096 and dry12-anchor-131072 under 4G (kept as superseded receipts)", "ruled_by": "the lane, under PLAN-v2 Appendix A/B item 14 ('raise to a measured value if it OOMs; never remove')", "utc": "2026-09-23T18:31Z", "what": "the bench instance's MemoryMax (Appendix A: 4G, marked UNTESTED)" }, { "id": "A-L03", "new": "MemoryMax=32G (applied live: systemctl --user set-property)", "old": "MemoryMax=16G (A-L02)", "reason": "16G still bound: llama-server's host prompt cache grows to its own 8,192 MiB limit (its log: '54 prompts, 8139.974 MiB (limits: 8192.000 MiB ...)'), and with an ~18 GB model file's page cache charged to the same cgroup the cap was hit 5,598 times during k0a-gemma4-anchor-cd and 6,994 times during b-vicuna-33b; both arms show 1.4-7.2 s prefill stalls on 1-3 items that benchbox's same arms do not. Anon stays <= ~9 GB; the rest is reclaimable page cache, so 32G is a ceiling that should not bind, and every arm.json carries the cgroup's memory.events before/after to prove it", "rows_the_author_had_seen": "k0a-gemma4-anchor-cd, k1-mistral-v0.1-cd, b-falcon-7b, b-vicuna-33b, and b-llama2-7b's task c", "ruled_by": "the lane, under PLAN-v2 Appendix B item 14 (as A-L02)", "utc": "2026-09-23T18:51:52Z", "what": "the bench instance's MemoryMax" }, { "id": "A-L04", "new": "nvidia-powerd active, clocks.max.sm 3090 (lock off), AC online, one board, and enforced.power.limit anywhere in the Dynamic Boost envelope (> 95 W and <= 175 W); the limit is reported per task as min/median/max from the 1 Hz ext trace, never as one number", "old": "enforced.power.limit == 150.00 W (+-0.5), clocks.max.sm 3090, powerd active, AC on, one board", "reason": "under Dynamic Boost the firmware moves the GPU's enforced limit when the CPU draws: the 1 Hz traces read 135.00-150.00 W on b-falcon-7b and b-vicuna-33b and 137.08-150.00 W on b-llama2-7b (a steady 150.00 on the anchor and mistral; loadavg ~5 from peers); the old gate refused b-llama2-7b before its task d at 18:51:55Z (146.76 W). Memory laptop-5090-refuses-a-power-cap: the boost posture's cap is 'a range from the trace'", "rows_the_author_had_seen": "the same as A-L03", "ruled_by": "the lead under R-0 (message to the laptop lane, received 18:53Z)", "utc": "2026-09-23T18:53Z", "what": "the laptop posture gate (PLAN-v2 §2.2: 'enforced.power.limit 150.00 W ... gated per arm, rc 4 on mismatch')" } ], "argv": [ "coilaptop.py", "--arm-id", "b-qwen3.6-27b", "--out", "/workshop/estate/bench/cost-of-intelligence-2026-09-23/laptop-reading-0/rows/b-qwen3.6-27b", "--model", "qwen3.6:27b-q4_K_M", "--num-ctx", "4096", "--think", "off", "--think-budget", "4096", "--tasks", "c,d", "--probes" ], "arm_id": "b-qwen3.6-27b", "baseline_rows": [ { "memory_mib": 15.0, "sm_mhz": 1192.0, "temp_c": 34.0, "uuid": "", "watts": 11.32 } ], "board": "", "box": "the laptop", "box_class": "gpu-5090-laptop-24g", "digest_pin": "3a40c32f1450", "disk_floor": { "df": "Avail\n113724358656", "floor_gib": 60.0, "free_gib": 105.9140625, "store": "/workshop/bench-coi-2026-09-23/store" }, "ended": "2026-09-23T21:36:54Z", "energy_instruments": [ "board (nvidia-smi power.draw by UUID, 1 Hz; run.py Boards)", "rapl psys + package-0 (+ core, uncore) via /run/rapl/energy (R-10d); not wall" ], "host": "http://127.0.0.1:withheld-2026-10-14", "idle_preload": { "board": { "cards": [ "" ], "joules": 159.47607882976533, "mean_watts": 11.384643061275517, "peak_watts": 11.44, "per_card": { "": { "joules": 159.47607882976533, "mean_watts": 11.384643061275517, "min_watts": 11.33, "peak_watts": 11.44, "samples": 15, "seconds": 14.007999897003174 } }, "samples": 15, "seconds": 14.007999897003174 }, "rapl": { "domains": { "core": { "joules": 99.538991, "mean_watts": 7.430179544724997, "samples": 14, "seconds": 13.396579504013062, "wraps": 0 }, "package-0": { "joules": 215.97369, "mean_watts": 16.121338710707654, "samples": 14, "seconds": 13.396759033203125, "wraps": 0 }, "psys": { "joules": 3.510611, "mean_watts": 0.2620624696471447, "samples": 14, "seconds": 13.396084547042847, "wraps": 0 }, "uncore": { "joules": 0.483824, "mean_watts": 0.03611601265981248, "samples": 14, "seconds": 13.396384716033936, "wraps": 0 } }, "instrument": "RAPL via /run/rapl/energy (root sampler, 1 Hz; R-10d): psys = the platform's own reading (not wall); package-0 = the CPU package" }, "window_epoch": [ 1790199202.539, 1790199216.547 ] }, "instrument": { "coilaptop_py": "0b5dcb41ac989a57dfe4912740dfd24ccd600da5ba3fbc181688d18a8adf94f0", "jev/kit/doorman_planted.json": "4daf059c26a47a84fa3b73630a64ba1a348da72c2e5df7f3c28c007b67da22f7", "jev/kit/task_c.json": "61a0b1d6e7cf66fd91992c0c3a4572f6fbae90e6471048fd7dc1eb41ff2d7f39", "jev/run.py": "withheld-2026-10-14", "jev/run_addenda.py": "a34aca61833ed0c5860c91c7e2e227429e8ea9e2e50fec68d7b7fec15d89acab" }, "limit": null, "memory_gate": { "amendment": "A-L01", "limits": { "mem_available_gib_min": 10.0, "psi_full_avg60_max_pct": 10.0, "shm_used_gib_max": 4.0, "swap_used_frac_max": 0.8 }, "mem_available_gib": 42.634342193603516, "psi_memory": { "full": { "avg10": 0.0, "avg300": 0.11, "avg60": 0.0, "total": 118579083.0 }, "some": { "avg10": 0.0, "avg300": 0.12, "avg60": 0.0, "total": 126469264.0 } }, "shm_used_gib": 2.8118934631347656, "stamp": "2026-09-23T21:33:16Z", "swap_total_gib": 17.99999237060547, "swap_used_frac": 0.342, "swaps": [ "Filename\t\t\t\tType\t\tSize\t\tUsed\t\tPriority", "/swap.img file\t\t2097148\t\t59120\t\t-1", "/dev/zram0 partition\t16777212\t6396436\t\t100" ], "why": "ok" }, "model": "qwen3.6:27b-q4_K_M", "num_ctx": 4096, "ollama_version": "0.32.13", "pin": { "api_ps_model": { "context_length": 4096, "details": { "families": [ "qwen35" ], "family": "qwen35", "format": "gguf", "parameter_size": "27.8B", "parent_model": "", "quantization_level": "Q4_K_M" }, "digest": "3a40c32f1450b8380412898385b0e00df5d6d2d801dd192ca1acb7e735cd050e", "expires_at": "2319-01-03T21:20:59.251829078Z", "model": "qwen3.6:27b-q4_K_M", "name": "qwen3.6:27b-q4_K_M", "size": 16325279742, "size_vram": 16325279742 }, "card": "", "cards": [ "" ], "fit": { "growth_bytes": 17727225856.0, "ps_size": 16325279742, "ps_size_vram": 16325279742, "ps_whole": true, "rule": "WHOLE when the named board grew by at least 90 %% of the store's bytes for this tag (weights + projector/draft; KV rides on top) AND /api/ps does not report a CPU share; ps alone is corroboration, never the verdict", "store_bytes": 17420432475, "whole": true }, "floor_mib": 200.0, "growth_mib": { "": 16906.0 }, "memory_after_mib": { "": 16921.0 }, "named_grew": true, "ok": true, "others_quiet": true, "per_card_grew": { "": true } }, "postflight": { "after_unload": { "api_ps": { "models": [] }, "api_ps_empty": true, "back_to_baseline": true, "board_mib_baseline": 15.0, "board_mib_now": 15.0, "rows": [ { "memory_mib": 15.0, "sm_mhz": 1192.0, "temp_c": 45.0, "uuid": "", "watts": 16.27 } ], "stamp": "2026-09-23T21:36:54Z" }, "posture_after": { "ac_online": "1", "battery": { "capacity": "93", "status": "Not charging" }, "boards_on_bus": [ "0, , NVIDIA GeForce RTX 5090 Laptop GPU, withheld-2026-10-14" ], "card": { "clocks.max.sm": "3090", "clocks.mem": "14001", "clocks.sm": "1935", "clocks_throttle_reasons.active": "0x0000000000000000", "driver_version": "595.91.07", "enforced.power.limit": "150.00", "memory.total": "24463", "memory.used": "16929", "name": "NVIDIA GeForce RTX 5090 Laptop GPU", "pci.bus_id": "withheld-2026-10-14", "pcie.link.gen.current": "5", "pcie.link.gen.max": "5", "pcie.link.width.current": "8", "pcie.link.width.max": "16", "persistence_mode": "Enabled", "power.default_limit": "95.00", "power.draw": "104.27", "power.limit": "[N/A]", "power.max_limit": "175.00", "pstate": "P0", "temperature.gpu": "55", "utilization.gpu": "0", "uuid": "", "vbios_version": "98.03.A2.00.21" }, "compute_apps": [ ", 214508, /workshop/bench-coi-2026-09-23/ollama-0.32.13/lib/ollama/llama-server, 16906 MiB" ], "fan": "unreadable on this board (fan.speed N/A); the operator's turbo fan per OPERATOR-CONDITIONS.md", "how": "nvidia-smi -i ; enforced.power.limit and clocks.max.sm READ, never typed; power.limit reads N/A on this board and the board refuses -pl", "loadavg": "2.70 2.73 2.70 1/1880 232911", "nvidia_powerd": "active", "stamp": "2026-09-23T21:36:48Z" }, "seat_after": { "api_ps": { "models": [] }, "stamp": "2026-09-23T21:36:54Z" }, "serve_cgroup_memory_after": { "events": { "high": "0", "low": "0", "max": "43078", "oom": "0", "oom_group_kill": "0", "oom_kill": "0", "sock_throttled": "401" }, "pressure": [ "some avg10=0.00 avg60=0.00 avg300=0.00 total=15976543", "full avg10=0.00 avg60=0.00 avg300=0.00 total=15976543" ], "stat": { "anon": 62681088, "file": 16928198656, "file_mapped": 1757184, "shmem": 0 }, "swap_current": 2600960 }, "serve_unit_after": { "ActiveState": "active", "ControlGroup": "/user.slice/user-1000.slice/user@1000.service/app.slice/coi-ollama-laptop.service", "Environment": "TZ=UTC OLLAMA_HOST=127.0.0.1:withheld-2026-10-14 OLLAMA_KEEP_ALIVE=-1 OLLAMA_MAX_LOADED_MODELS=1 OLLAMA_=1 OLLAMA_FLASH_ATTENTION=1 OLLAMA_KV_CACHE_TYPE=f16 OLLAMA_NOPRUNE=1 OLLAMA_MODELS=/workshop/bench-coi-2026-09-23/store CUDA_VISIBLE_DEVICES= OLLAMA_VULKAN=0 GGML_VK_VISIBLE_DEVICES=-1 HIP_VISIBLE_DEVICES= ROCR_VISIBLE_DEVICES=", "ExecStart": "{ path=/workshop/bench-coi-2026-09-23/ollama-0.32.13/bin/ollama ; argv[]=/workshop/bench-coi-2026-09-23/ollama-0.32.13/bin/ollama serve ; ignore_errors=no ; start_time=[n/a] ; stop_time=[n/a] ; pid=0 ; code=(null) ; status=0/0 }", "MainPID": "1739347", "MemoryCurrent": "17477799936", "MemoryMax": "34359738368", "MemoryPeak": "30671613952", "NRestarts": "0" } }, "posture_before": { "ac_online": "1", "battery": { "capacity": "93", "status": "Not charging" }, "boards_on_bus": [ "0, , NVIDIA GeForce RTX 5090 Laptop GPU, withheld-2026-10-14" ], "card": { "clocks.max.sm": "3090", "clocks.mem": "810", "clocks.sm": "1192", "clocks_throttle_reasons.active": "0x0000000000000000", "driver_version": "595.91.07", "enforced.power.limit": "150.00", "memory.total": "24463", "memory.used": "15", "name": "NVIDIA GeForce RTX 5090 Laptop GPU", "pci.bus_id": "withheld-2026-10-14", "pcie.link.gen.current": "2", "pcie.link.gen.max": "5", "pcie.link.width.current": "8", "pcie.link.width.max": "16", "persistence_mode": "Enabled", "power.default_limit": "95.00", "power.draw": "11.31", "power.limit": "[N/A]", "power.max_limit": "175.00", "pstate": "P5", "temperature.gpu": "35", "utilization.gpu": "0", "uuid": "", "vbios_version": "98.03.A2.00.21" }, "compute_apps": [], "fan": "unreadable on this board (fan.speed N/A); the operator's turbo fan per OPERATOR-CONDITIONS.md", "how": "nvidia-smi -i ; enforced.power.limit and clocks.max.sm READ, never typed; power.limit reads N/A on this board and the board refuses -pl", "loadavg": "2.37 2.62 2.68 2/1850 211128", "nvidia_powerd": "active", "stamp": "2026-09-23T21:33:16Z" }, "posture_verdict": "Dynamic Boost: enforced 150.00 W (inside the (95, 175] W envelope), clock lock off (3090 MHz), powerd active, AC on, one board", "probes": { "decode": { "board": { "cards": [ "" ], "joules": 2812.896460145712, "mean_watts": 127.78922630479798, "peak_watts": 140.36, "per_card": { "": { "joules": 2812.896460145712, "mean_watts": 127.78922630479798, "min_watts": 23.28, "peak_watts": 140.36, "samples": 23, "seconds": 22.01200008392334 } }, "samples": 23, "seconds": 22.01200008392334 }, "num_predict": 256, "prompt": "Describe, in about three hundred words, a lighthouse on a rocky coast during a winter storm.", "rapl": { "domains": { "core": { "joules": 432.158378, "mean_watts": 20.956425168402706, "samples": 21, "seconds": 20.621760368347168, "wraps": 0 }, "package-0": { "joules": 576.016103, "mean_watts": 27.93239150257861, "samples": 21, "seconds": 20.62179684638977, "wraps": 0 }, "psys": { "joules": 5.519884, "mean_watts": 0.2676671482440994, "samples": 21, "seconds": 20.622194528579712, "wraps": 0 }, "uncore": { "joules": 0.763426, "mean_watts": 0.037019703070195, "samples": 21, "seconds": 20.622153520584106, "wraps": 0 } }, "instrument": "RAPL via /run/rapl/energy (root sampler, 1 Hz; R-10d): psys = the platform's own reading (not wall); package-0 = the CPU package" }, "runs": [ { "decode_tok_s": 37.979524882078024, "done_reason": "length", "eval_count": 256, "eval_s": 6.740474, "load_s": 0.192231446, "prefill_tok_s": 168.98760581528597, "prompt_eval_count": 32, "prompt_eval_s": 0.189363, "runner_pids": [ 214508, 1739347 ], "text_head": "The sky was a bruised purple, churning with the violent energy of a winter gale that had been battering the coast for ho", "thinking_chars": 0, "total_s": 7.343121816 }, { "decode_tok_s": 38.01444608348744, "done_reason": "length", "eval_count": 256, "eval_s": 6.734282, "load_s": 0.201897335, "prefill_tok_s": 305.6059593162067, "prompt_eval_count": 32, "prompt_eval_s": 0.10471, "runner_pids": [ 214508, 1739347 ], "text_head": "The sky was a bruised purple, churning with the violent energy of a winter gale that had been battering the coast for ho", "thinking_chars": 0, "total_s": 7.245597154 }, { "decode_tok_s": 37.95201442176548, "done_reason": "length", "eval_count": 256, "eval_s": 6.74536, "load_s": 0.214033641, "prefill_tok_s": 294.99612817581766, "prompt_eval_count": 32, "prompt_eval_s": 0.108476, "runner_pids": [ 214508, 1739347 ], "text_head": "The sky was a bruised purple, churning with the violent energy of a winter gale that had been battering the coast for ho", "thinking_chars": 0, "total_s": 7.070949687 } ], "window_epoch": [ 1790199361.454, 1790199383.466 ] }, "prefill": { "board": { "cards": [ "" ], "joules": 1537.9133232712745, "mean_watts": 118.1102322467237, "peak_watts": 154.96, "per_card": { "": { "joules": 1537.9133232712745, "mean_watts": 118.1102322467237, "min_watts": 39.6, "peak_watts": 154.96, "samples": 14, "seconds": 13.020999908447266 } }, "samples": 14, "seconds": 13.020999908447266 }, "rapl": { "domains": { "core": { "joules": 228.11105, "mean_watts": 18.415395581985397, "samples": 13, "seconds": 12.386975288391113, "wraps": 0 }, "package-0": { "joules": 318.445352, "mean_watts": 25.709020212850888, "samples": 13, "seconds": 12.38652229309082, "wraps": 0 }, "psys": { "joules": 3.348869, "mean_watts": 0.2703593213210032, "samples": 13, "seconds": 12.386734008789062, "wraps": 0 }, "uncore": { "joules": 0.484801, "mean_watts": 0.03913849566263718, "samples": 13, "seconds": 12.386807203292847, "wraps": 0 } }, "instrument": "RAPL via /run/rapl/energy (root sampler, 1 Hz; R-10d): psys = the platform's own reading (not wall); package-0 = the CPU package" }, "runs": [ { "decode_tok_s": 1000000.0, "done_reason": "length", "eval_count": 1, "eval_s": 1e-06, "load_s": 0.201115169, "prefill_tok_s": 1437.4560521473231, "prompt_eval_count": 2263, "prompt_eval_s": 1.574309, "runner_pids": [ 214508, 1739347 ], "text_head": "Dar", "thinking_chars": 0, "total_s": 1.789492055 }, { "decode_tok_s": 1000000.0, "done_reason": "length", "eval_count": 1, "eval_s": 1e-06, "load_s": 0.204471083, "prefill_tok_s": 1434.1523626724584, "prompt_eval_count": 2261, "prompt_eval_s": 1.576541, "runner_pids": [ 214508, 1739347 ], "text_head": "Dar", "thinking_chars": 0, "total_s": 2.071113507 }, { "decode_tok_s": 1000000.0, "done_reason": "length", "eval_count": 1, "eval_s": 1e-06, "load_s": 0.208320944, "prefill_tok_s": 1428.2620349387496, "prompt_eval_count": 2262, "prompt_eval_s": 1.583743, "runner_pids": [ 214508, 1739347 ], "text_head": "Dar", "thinking_chars": 0, "total_s": 2.067105723 }, { "decode_tok_s": 1000000.0, "done_reason": "length", "eval_count": 1, "eval_s": 1e-06, "load_s": 0.203216646, "prefill_tok_s": 1428.395440000758, "prompt_eval_count": 2261, "prompt_eval_s": 1.582895, "runner_pids": [ 214508, 1739347 ], "text_head": "Dar", "thinking_chars": 0, "total_s": 2.085308701 }, { "decode_tok_s": 1000000.0, "done_reason": "length", "eval_count": 1, "eval_s": 1e-06, "load_s": 0.202596904, "prefill_tok_s": 1432.2677784033756, "prompt_eval_count": 2264, "prompt_eval_s": 1.58071, "runner_pids": [ 214508, 1739347 ], "text_head": "Dar", "thinking_chars": 0, "total_s": 2.069607223 }, { "decode_tok_s": 1000000.0, "done_reason": "length", "eval_count": 1, "eval_s": 1e-06, "load_s": 0.156339724, "prefill_tok_s": 1431.621534781939, "prompt_eval_count": 2259, "prompt_eval_s": 1.577931, "runner_pids": [ 214508, 1739347 ], "text_head": "Dar", "thinking_chars": 0, "total_s": 2.01080373 } ], "target_tokens_approx": 3000.0, "window_epoch": [ 1790199384.519, 1790199397.54 ] }, "ttft": { "how": "the first 12 task_c prompts, run.py's body with stream true, in task order (so the prompt-cache condition is the task's own); TTFT = client clock on localhost from the request to the first chunk carrying content or thinking", "median_s": 0.7425450800001272, "p95_s": 0.7665347369911615, "runs": [ { "eval_count": 3, "prompt_eval_count": 228, "server_load_ms": 187.359166, "server_prompt_eval_ms": 268.537, "total_s": 0.7970713090035133, "ttft_s": 0.7396387000044342 }, { "eval_count": 3, "prompt_eval_count": 324, "server_load_ms": 172.271974, "server_prompt_eval_ms": 341.219, "total_s": 0.7856577320053475, "ttft_s": 0.7290144809958292 }, { "eval_count": 3, "prompt_eval_count": 245, "server_load_ms": 212.303623, "server_prompt_eval_ms": 283.503, "total_s": 0.7800548810046166, "ttft_s": 0.7234846250066767 }, { "eval_count": 3, "prompt_eval_count": 346, "server_load_ms": 173.703712, "server_prompt_eval_ms": 354.306, "total_s": 0.8012593720050063, "ttft_s": 0.7454514599958202 }, { "eval_count": 3, "prompt_eval_count": 356, "server_load_ms": 154.982104, "server_prompt_eval_ms": 359.008, "total_s": 0.7874042949988507, "ttft_s": 0.7323194719938328 }, { "eval_count": 3, "prompt_eval_count": 358, "server_load_ms": 188.001854, "server_prompt_eval_ms": 357.776, "total_s": 0.8227802059991518, "ttft_s": 0.7665347369911615 }, { "eval_count": 3, "prompt_eval_count": 263, "server_load_ms": 162.355479, "server_prompt_eval_ms": 309.473, "total_s": 0.7413623039901722, "ttft_s": 0.6862170709937345 }, { "eval_count": 3, "prompt_eval_count": 244, "server_load_ms": 182.46816, "server_prompt_eval_ms": 286.94, "total_s": 0.7454578049946576, "ttft_s": 0.6890791640034877 }, { "eval_count": 3, "prompt_eval_count": 320, "server_load_ms": 190.685308, "server_prompt_eval_ms": 337.666, "total_s": 0.8052883659984218, "ttft_s": 0.7483459529903485 }, { "eval_count": 3, "prompt_eval_count": 354, "server_load_ms": 165.998273, "server_prompt_eval_ms": 355.192, "total_s": 0.8090219399891794, "ttft_s": 0.7534447799989721 }, { "eval_count": 3, "prompt_eval_count": 347, "server_load_ms": 180.107439, "server_prompt_eval_ms": 358.089, "total_s": 0.8275919840089045, "ttft_s": 0.7728563800046686 }, { "eval_count": 3, "prompt_eval_count": 349, "server_load_ms": 180.248501, "server_prompt_eval_ms": 354.793, "total_s": 0.8196494830044685, "ttft_s": 0.7645935020118486 } ] } }, "rapl_probe": [ "1790199195.835700717 package-0 103172600257 262143328850", "1790199195.843502031 core 210650410135 262143328850", "1790199195.850228216 uncore 7684045087 262143328850", "1790199195.856847670 psys 22867947927 262143328850" ], "rapl_unreadable_polls": 0, "seat_before": { "api_ps": { "models": [] }, "stamp": "2026-09-23T21:33:16Z" }, "serve_cgroup_memory_before": { "events": { "high": "0", "low": "0", "max": "43078", "oom": "0", "oom_group_kill": "0", "oom_kill": "0", "sock_throttled": "401" }, "pressure": [ "some avg10=0.00 avg60=0.00 avg300=0.00 total=15976543", "full avg10=0.00 avg60=0.00 avg300=0.00 total=15976543" ], "stat": { "anon": 30298112, "file": 18181066752, "file_mapped": 1757184, "shmem": 0 }, "swap_current": 2617344 }, "serve_unit_before": { "ActiveState": "active", "ControlGroup": "/user.slice/user-1000.slice/user@1000.service/app.slice/coi-ollama-laptop.service", "Environment": "TZ=UTC OLLAMA_HOST=127.0.0.1:withheld-2026-10-14 OLLAMA_KEEP_ALIVE=-1 OLLAMA_MAX_LOADED_MODELS=1 OLLAMA_=1 OLLAMA_FLASH_ATTENTION=1 OLLAMA_KV_CACHE_TYPE=f16 OLLAMA_NOPRUNE=1 OLLAMA_MODELS=/workshop/bench-coi-2026-09-23/store CUDA_VISIBLE_DEVICES= OLLAMA_VULKAN=0 GGML_VK_VISIBLE_DEVICES=-1 HIP_VISIBLE_DEVICES= ROCR_VISIBLE_DEVICES=", "ExecStart": "{ path=/workshop/bench-coi-2026-09-23/ollama-0.32.13/bin/ollama ; argv[]=/workshop/bench-coi-2026-09-23/ollama-0.32.13/bin/ollama serve ; ignore_errors=no ; start_time=[n/a] ; stop_time=[n/a] ; pid=0 ; code=(null) ; status=0/0 }", "MainPID": "1739347", "MemoryCurrent": "18732216320", "MemoryMax": "34359738368", "MemoryPeak": "30671613952", "NRestarts": "0" }, "show": { "capabilities": [ "completion", "vision", "tools", "thinking" ], "details": { "families": [ "qwen35" ], "family": "qwen35", "format": "gguf", "parameter_size": "27.8B", "parent_model": "qwen3.6-source:27b-q4_K_M-20260824", "quantization_level": "Q4_K_M" }, "model_info": { "general.architecture": "qwen35", "general.file_type": 15, "general.parameter_count": 27781427952, "general.quantization_version": 2, "qwen35.attention.head_count": 24, "qwen35.attention.head_count_kv": null, "qwen35.attention.key_length": 256, "qwen35.attention.layer_norm_rms_epsilon": 1e-06, "qwen35.attention.value_length": 256, "qwen35.block_count": 64, "qwen35.context_length": 262144, "qwen35.embedding_length": 5120, "qwen35.feed_forward_length": 17408, "qwen35.full_attention_interval": 4, "qwen35.image_token_id": 248056, "qwen35.mrope_sections": null, "qwen35.rope.dimension_count": 64, "qwen35.rope.dimension_sections": null, "qwen35.rope.freq_base": 10000000, "qwen35.rope.mrope_interleaved": true, "qwen35.rope.mrope_section": null, "qwen35.ssm.conv_kernel": 4, "qwen35.ssm.group_count": 16, "qwen35.ssm.inner_size": 6144, "qwen35.ssm.state_size": 128, "qwen35.ssm.time_step_rank": 48, "qwen35.ssm.v_head_reordered": true, "qwen35.vision.attention.head_count": 16, "qwen35.vision.block_count": 27, "qwen35.vision.embedding_length": 1152, "qwen35.vision.image_mean": null, "qwen35.vision.image_std": null, "qwen35.vision.longest_edge": 16777216, "qwen35.vision.num_channels": 3, "qwen35.vision.patch_size": 16, "qwen35.vision.shortest_edge": 65536, "qwen35.vision.spatial_merge_size": 2, "qwen35.vision.temporal_patch_size": 2, "qwen35.vision_end_token_id": 248054, "qwen35.vision_start_token_id": 248053, "tokenizer.ggml.add_eos_token": false, "tokenizer.ggml.add_padding_token": false, "tokenizer.ggml.eos_token_id": 248046, "tokenizer.ggml.eos_token_ids": null, "tokenizer.ggml.merges": null, "tokenizer.ggml.model": "gpt2", "tokenizer.ggml.padding_token_id": 248044, "tokenizer.ggml.pre": "qwen35", "tokenizer.ggml.scores": null, "tokenizer.ggml.token_type": null, "tokenizer.ggml.tokens": null }, "parameters": "min_p 0\npresence_penalty 0\nrepeat_penalty 1\ntemperature 1\ntop_k 20\ntop_p 0.95", "system_sha256": null, "template_head": "{{ .Prompt }}", "template_sha256": "b507b9c2f6ca642bffcd06665ea7c91f235fd32daeefdf875a0f938db05fb315" }, "started": "2026-09-23T21:33:16Z", "tag": { "details": { "context_length": 262144, "embedding_length": 5120, "families": [ "qwen35" ], "family": "qwen35", "format": "gguf", "parameter_size": "27.8B", "parent_model": "qwen3.6-source:27b-q4_K_M-20260824", "quantization_level": "Q4_K_M" }, "digest": "3a40c32f1450b8380412898385b0e00df5d6d2d801dd192ca1acb7e735cd050e", "modified_at": "2026-09-23T21:33:16.618890844Z", "name": "qwen3.6:27b-q4_K_M", "size": 17420432475 }, "tasks": [ "c", "d" ], "tasks_run": { "c": { "board_idle_window_epoch": [ 1790199225.524, 1790199234.54 ], "board_task_window_epoch": [ 1790199235.59, 1790199299.622 ], "energy": { "cards": [ "" ], "decisions": 108, "joules": 5817.824018068313, "joules_per_decision": 53.86874090803994, "mean_watts": 90.85807115467549, "net_joules_per_decision": 40.528404205368005, "peak_watts": 108.91, "per_card": { "": { "joules": 5817.824018068313, "mean_watts": 90.85807115467549, "min_watts": 22.49, "peak_watts": 108.91, "samples": 65, "seconds": 64.03200006484985 } }, "samples": 65, "seconds": 64.03200006484985 }, "idle": { "cards": [ "" ], "joules": 202.86512072682382, "mean_watts": 22.500567879020025, "peak_watts": 22.56, "per_card": { "": { "joules": 202.86512072682382, "mean_watts": 22.500567879020025, "min_watts": 22.47, "peak_watts": 22.56, "samples": 10, "seconds": 9.016000032424927 } }, "samples": 10, "seconds": 9.016000032424927, "what": "the draw of these cards immediately before the arm, with the model resident and no work in flight" }, "posture_end": { "ac_online": "1", "battery": { "capacity": "93", "status": "Not charging" }, "boards_on_bus": [ "0, , NVIDIA GeForce RTX 5090 Laptop GPU, withheld-2026-10-14" ], "card": { "clocks.max.sm": "3090", "clocks.mem": "14001", "clocks.sm": "1590", "clocks_throttle_reasons.active": "0x0000000000000000", "driver_version": "595.91.07", "enforced.power.limit": "150.00", "memory.total": "24463", "memory.used": "16929", "name": "NVIDIA GeForce RTX 5090 Laptop GPU", "pci.bus_id": "withheld-2026-10-14", "pcie.link.gen.current": "5", "pcie.link.gen.max": "5", "pcie.link.width.current": "8", "pcie.link.width.max": "16", "persistence_mode": "Enabled", "power.default_limit": "95.00", "power.draw": "32.44", "power.limit": "[N/A]", "power.max_limit": "175.00", "pstate": "P0", "temperature.gpu": "43", "utilization.gpu": "0", "uuid": "", "vbios_version": "98.03.A2.00.21" }, "compute_apps": [ ", 214508, /workshop/bench-coi-2026-09-23/ollama-0.32.13/lib/ollama/llama-server, 16906 MiB" ], "fan": "unreadable on this board (fan.speed N/A); the operator's turbo fan per OPERATOR-CONDITIONS.md", "how": "nvidia-smi -i ; enforced.power.limit and clocks.max.sm READ, never typed; power.limit reads N/A on this board and the board refuses -pl", "loadavg": "2.84 2.67 2.68 1/1873 220610", "nvidia_powerd": "active", "stamp": "2026-09-23T21:35:01Z" }, "posture_start": { "ac_online": "1", "battery": { "capacity": "93", "status": "Not charging" }, "boards_on_bus": [ "0, , NVIDIA GeForce RTX 5090 Laptop GPU, withheld-2026-10-14" ], "card": { "clocks.max.sm": "3090", "clocks.mem": "14001", "clocks.sm": "1590", "clocks_throttle_reasons.active": "0x0000000000000000", "driver_version": "595.91.07", "enforced.power.limit": "150.00", "memory.total": "24463", "memory.used": "16921", "name": "NVIDIA GeForce RTX 5090 Laptop GPU", "pci.bus_id": "withheld-2026-10-14", "pcie.link.gen.current": "5", "pcie.link.gen.max": "5", "pcie.link.width.current": "8", "pcie.link.width.max": "16", "persistence_mode": "Enabled", "power.default_limit": "95.00", "power.draw": "22.48", "power.limit": "[N/A]", "power.max_limit": "175.00", "pstate": "P0", "temperature.gpu": "35", "utilization.gpu": "0", "uuid": "", "vbios_version": "98.03.A2.00.21" }, "compute_apps": [ ", 214508, /workshop/bench-coi-2026-09-23/ollama-0.32.13/lib/ollama/llama-server, 16898 MiB" ], "fan": "unreadable on this board (fan.speed N/A); the operator's turbo fan per OPERATOR-CONDITIONS.md", "how": "nvidia-smi -i ; enforced.power.limit and clocks.max.sm READ, never typed; power.limit reads N/A on this board and the board refuses -pl", "loadavg": "2.12 2.54 2.65 1/1868 215205", "nvidia_powerd": "active", "stamp": "2026-09-23T21:33:45Z" }, "rapl_idle_resident": { "domains": { "core": { "joules": 39.599569, "mean_watts": 4.797861978277633, "samples": 9, "seconds": 8.253586530685425, "wraps": 0 }, "package-0": { "joules": 100.097034, "mean_watts": 12.126751162809, "samples": 9, "seconds": 8.254233360290527, "wraps": 0 }, "psys": { "joules": 2.159235, "mean_watts": 0.26163767744245153, "samples": 9, "seconds": 8.25276780128479, "wraps": 0 }, "uncore": { "joules": 0.333068, "mean_watts": 0.04035723948051107, "samples": 9, "seconds": 8.252992630004883, "wraps": 0 } }, "instrument": "RAPL via /run/rapl/energy (root sampler, 1 Hz; R-10d): psys = the platform's own reading (not wall); package-0 = the CPU package" }, "rapl_task": { "domains": { "core": { "joules": 1119.91205, "mean_watts": 17.792846791124568, "samples": 62, "seconds": 62.94170141220093, "wraps": 0 }, "package-0": { "joules": 1623.229426, "mean_watts": 25.789787802335564, "samples": 62, "seconds": 62.94078254699707, "wraps": 0 }, "psys": { "joules": 16.613849, "mean_watts": 0.26395019320017643, "samples": 62, "seconds": 62.94312119483948, "wraps": 0 }, "uncore": { "joules": 2.259394, "mean_watts": 0.03589624478914815, "samples": 62, "seconds": 62.942349910736084, "wraps": 0 } }, "instrument": "RAPL via /run/rapl/energy (root sampler, 1 Hz; R-10d): psys = the platform's own reading (not wall); package-0 = the CPU package" }, "seat_requests_task": { "gin_lines": 0, "how": "journalctl -u ollama (the system daemon :withheld-2026-10-14), read-only; a GIN line is one request it served", "lines": [], "since": "2026-09-23 21:33:55 UTC", "until": "2026-09-23 21:35:00 UTC" }, "summary": { "accuracy": 0.8796296296296297, "brier": 0.24074074074074073, "correct": 95, "errors": 0, "eval_tokens_total": 291, "floored_rows": 0, "letters_floored_total": 0, "median_seconds": 0.5895324015000369, "n": 108, "n_labelled": 108, "n_reached": 108, "p95_seconds": 0.648852293001255, "prompt_tokens_total": 31303, "refused": 0, "refused_by_reason": {}, "reliability": [ { "accuracy": null, "correct": 0, "gap": null, "hi": 0.1, "lo": 0.0, "mean_p": null, "n": 0, "sum_p": 0.0 }, { "accuracy": null, "correct": 0, "gap": null, "hi": 0.2, "lo": 0.1, "mean_p": null, "n": 0, "sum_p": 0.0 }, { "accuracy": null, "correct": 0, "gap": null, "hi": 0.3, "lo": 0.2, "mean_p": null, "n": 0, "sum_p": 0.0 }, { "accuracy": null, "correct": 0, "gap": null, "hi": 0.4, "lo": 0.3, "mean_p": null, "n": 0, "sum_p": 0.0 }, { "accuracy": null, "correct": 0, "gap": null, "hi": 0.5, "lo": 0.4, "mean_p": null, "n": 0, "sum_p": 0.0 }, { "accuracy": null, "correct": 0, "gap": null, "hi": 0.6, "lo": 0.5, "mean_p": null, "n": 0, "sum_p": 0.0 }, { "accuracy": null, "correct": 0, "gap": null, "hi": 0.7, "lo": 0.6, "mean_p": null, "n": 0, "sum_p": 0.0 }, { "accuracy": null, "correct": 0, "gap": null, "hi": 0.8, "lo": 0.7, "mean_p": null, "n": 0, "sum_p": 0.0 }, { "accuracy": null, "correct": 0, "gap": null, "hi": 0.9, "lo": 0.8, "mean_p": null, "n": 0, "sum_p": 0.0 }, { "accuracy": 0.8796296296296297, "correct": 95, "gap": 0.12037037037037035, "hi": 1.0, "lo": 0.9, "mean_p": 1.0, "n": 108, "sum_p": 108.0 } ], "unparsed": 0 }, "wall_seconds": 64.06680676399264 }, "d": { "board_idle_window_epoch": [ 1790199301.753, 1790199310.758 ], "board_task_window_epoch": [ 1790199311.786, 1790199358.712 ], "energy": { "cards": [ "" ], "decisions": 48, "joules": 4005.643204088211, "joules_per_decision": 83.45090008517106, "mean_watts": 85.36084930199513, "net_joules_per_decision": 60.81068150256545, "peak_watts": 104.59, "per_card": { "": { "joules": 4005.643204088211, "mean_watts": 85.36084930199513, "min_watts": 22.96, "peak_watts": 104.59, "samples": 44, "seconds": 46.925999879837036 } }, "samples": 44, "seconds": 46.925999879837036 }, "idle": { "cards": [ "" ], "joules": 208.54127712726594, "mean_watts": 23.15838756228636, "peak_watts": 23.34, "per_card": { "": { "joules": 208.54127712726594, "mean_watts": 23.15838756228636, "min_watts": 22.95, "peak_watts": 23.34, "samples": 10, "seconds": 9.004999876022339 } }, "samples": 10, "seconds": 9.004999876022339, "what": "the draw of these cards immediately before the arm, with the model resident and no work in flight" }, "posture_end": { "ac_online": "1", "battery": { "capacity": "93", "status": "Not charging" }, "boards_on_bus": [ "0, , NVIDIA GeForce RTX 5090 Laptop GPU, withheld-2026-10-14" ], "card": { "clocks.max.sm": "3090", "clocks.mem": "14001", "clocks.sm": "1590", "clocks_throttle_reasons.active": "0x0000000000000000", "driver_version": "595.91.07", "enforced.power.limit": "150.00", "memory.total": "24463", "memory.used": "16929", "name": "NVIDIA GeForce RTX 5090 Laptop GPU", "pci.bus_id": "withheld-2026-10-14", "pcie.link.gen.current": "5", "pcie.link.gen.max": "5", "pcie.link.width.current": "8", "pcie.link.width.max": "16", "persistence_mode": "Enabled", "power.default_limit": "95.00", "power.draw": "42.72", "power.limit": "[N/A]", "power.max_limit": "175.00", "pstate": "P0", "temperature.gpu": "45", "utilization.gpu": "0", "uuid": "", "vbios_version": "98.03.A2.00.21" }, "compute_apps": [ ", 214508, /workshop/bench-coi-2026-09-23/ollama-0.32.13/lib/ollama/llama-server, 16906 MiB" ], "fan": "unreadable on this board (fan.speed N/A); the operator's turbo fan per OPERATOR-CONDITIONS.md", "how": "nvidia-smi -i ; enforced.power.limit and clocks.max.sm READ, never typed; power.limit reads N/A on this board and the board refuses -pl", "loadavg": "3.73 2.88 2.75 2/1878 229097", "nvidia_powerd": "active", "stamp": "2026-09-23T21:36:00Z" }, "posture_start": { "ac_online": "1", "battery": { "capacity": "93", "status": "Not charging" }, "boards_on_bus": [ "0, , NVIDIA GeForce RTX 5090 Laptop GPU, withheld-2026-10-14" ], "card": { "clocks.max.sm": "3090", "clocks.mem": "14001", "clocks.sm": "1590", "clocks_throttle_reasons.active": "0x0000000000000000", "driver_version": "595.91.07", "enforced.power.limit": "150.00", "memory.total": "24463", "memory.used": "16929", "name": "NVIDIA GeForce RTX 5090 Laptop GPU", "pci.bus_id": "withheld-2026-10-14", "pcie.link.gen.current": "5", "pcie.link.gen.max": "5", "pcie.link.width.current": "8", "pcie.link.width.max": "16", "persistence_mode": "Enabled", "power.default_limit": "95.00", "power.draw": "23.30", "power.limit": "[N/A]", "power.max_limit": "175.00", "pstate": "P0", "temperature.gpu": "43", "utilization.gpu": "0", "uuid": "", "vbios_version": "98.03.A2.00.21" }, "compute_apps": [ ", 214508, /workshop/bench-coi-2026-09-23/ollama-0.32.13/lib/ollama/llama-server, 16906 MiB" ], "fan": "unreadable on this board (fan.speed N/A); the operator's turbo fan per OPERATOR-CONDITIONS.md", "how": "nvidia-smi -i ; enforced.power.limit and clocks.max.sm READ, never typed; power.limit reads N/A on this board and the board refuses -pl", "loadavg": "2.84 2.67 2.68 8/1873 220618", "nvidia_powerd": "active", "stamp": "2026-09-23T21:35:01Z" }, "rapl_idle_resident": { "domains": { "core": { "joules": 34.433322, "mean_watts": 4.175289893327605, "samples": 9, "seconds": 8.24692964553833, "wraps": 0 }, "package-0": { "joules": 98.979239, "mean_watts": 12.002237514207602, "samples": 9, "seconds": 8.246732234954834, "wraps": 0 }, "psys": { "joules": 2.162348, "mean_watts": 0.2621965794144831, "samples": 9, "seconds": 8.24704885482788, "wraps": 0 }, "uncore": { "joules": 0.358764, "mean_watts": 0.043502534389159506, "samples": 9, "seconds": 8.246967792510986, "wraps": 0 } }, "instrument": "RAPL via /run/rapl/energy (root sampler, 1 Hz; R-10d): psys = the platform's own reading (not wall); package-0 = the CPU package" }, "rapl_task": { "domains": { "core": { "joules": 996.785717, "mean_watts": 21.977409490271686, "samples": 42, "seconds": 45.35501408576965, "wraps": 0 }, "package-0": { "joules": 1382.139894, "mean_watts": 30.47405303803558, "samples": 42, "seconds": 45.3546462059021, "wraps": 0 }, "psys": { "joules": 12.027313, "mean_watts": 0.26517792014015146, "samples": 42, "seconds": 45.35563516616821, "wraps": 0 }, "uncore": { "joules": 1.681636, "mean_watts": 0.03707695987757047, "samples": 42, "seconds": 45.3552827835083, "wraps": 0 } }, "instrument": "RAPL via /run/rapl/energy (root sampler, 1 Hz; R-10d): psys = the platform's own reading (not wall); package-0 = the CPU package" }, "seat_requests_task": { "gin_lines": 0, "how": "journalctl -u ollama (the system daemon :withheld-2026-10-14), read-only; a GIN line is one request it served", "lines": [], "since": "2026-09-23 21:35:11 UTC", "until": "2026-09-23 21:35:59 UTC" }, "summary": { "accuracy": 0.6666666666666666, "brier": 0.6666666666666666, "correct": 32, "errors": 0, "eval_tokens_total": 144, "floored_rows": 0, "letters_floored_total": 0, "median_seconds": 0.9050825800004532, "n": 48, "n_labelled": 48, "n_reached": 48, "p95_seconds": 0.935071574000176, "prompt_tokens_total": 24550, "refused": 0, "refused_by_reason": {}, "reliability": [ { "accuracy": null, "correct": 0, "gap": null, "hi": 0.1, "lo": 0.0, "mean_p": null, "n": 0, "sum_p": 0.0 }, { "accuracy": null, "correct": 0, "gap": null, "hi": 0.2, "lo": 0.1, "mean_p": null, "n": 0, "sum_p": 0.0 }, { "accuracy": null, "correct": 0, "gap": null, "hi": 0.3, "lo": 0.2, "mean_p": null, "n": 0, "sum_p": 0.0 }, { "accuracy": null, "correct": 0, "gap": null, "hi": 0.4, "lo": 0.3, "mean_p": null, "n": 0, "sum_p": 0.0 }, { "accuracy": null, "correct": 0, "gap": null, "hi": 0.5, "lo": 0.4, "mean_p": null, "n": 0, "sum_p": 0.0 }, { "accuracy": null, "correct": 0, "gap": null, "hi": 0.6, "lo": 0.5, "mean_p": null, "n": 0, "sum_p": 0.0 }, { "accuracy": null, "correct": 0, "gap": null, "hi": 0.7, "lo": 0.6, "mean_p": null, "n": 0, "sum_p": 0.0 }, { "accuracy": null, "correct": 0, "gap": null, "hi": 0.8, "lo": 0.7, "mean_p": null, "n": 0, "sum_p": 0.0 }, { "accuracy": null, "correct": 0, "gap": null, "hi": 0.9, "lo": 0.8, "mean_p": null, "n": 0, "sum_p": 0.0 }, { "accuracy": 0.6666666666666666, "correct": 32, "gap": 0.33333333333333337, "hi": 1.0, "lo": 0.9, "mean_p": 1.0, "n": 48, "sum_p": 48.0 } ], "unparsed": 0 }, "wall_seconds": 47.05574912199518 } }, "temp_guard": { "max_temp_c": 59.0, "rule": ">= 87 C on 6 consecutive 5 s reads", "tripped": null }, "think": "off", "think_budget": 8, "util_before_pct": [ 0.0, 0.0, 0.0 ], "warm": { "done_reason": "length", "load_ms": 4671.238141, "runner_pids": [ 214508, 1739347 ], "seconds": 4.814948756000376 } }