← all runs · ← report

wi007_qwen3_vl_8b_instruct_wi006 / live_d2_none_s409 valid

What this trial was

runwi007_qwen3_vl_8b_instruct_wi006triallive_d2_none_s409conditionlivescenariopending_effectskinnotificationcanvas320×240 px (target 24 px)event at500 msresponse window16 sframe period16.67 msobservation publication delay30 msdeliverysimulated_fixed_delay, 20 msclockmonotonic_nseffect_delay_ns2000000000irrelevant_motionFalse

Controller: {"enable_thinking": false, "image_scale": 2, "invoke_on": "frame_change", "kind": "vlm_pending", "max_tokens": 64, "memory": "none", "model_id": "Qwen/Qwen3-VL-8B-Instruct", "pending_radius_px": 10, "pending_timeout_ns": 10000000000, "prompt_version": 4, "reasoning_effort": null, "temperature": 0.0}

Result for this trial

first click acceptedTruefirst E→A8.80 sduplicates0invalid clicks0suppressed proposals0waits16decisions17exact successTruedropped0contract violationsnone

Replay

key frames

Key frames: each decision's starting frame with the click it produced (amber = proposal before binding, red = committed, grey = suppressed), then the screen at each arrival A (green = valid).

The prompt sent with every frame

You control a desktop by clicking. You are shown one screenshot. At some point a notification card with a blue button appears at an unknown position and expires shortly after. If the card is visible, answer with action "click" and the coordinates of the center of its blue button on a 0-1000 grid (x: 0 = left edge, 1000 = right edge; y: 0 = top edge, 1000 = bottom edge). If no card is visible, answer with action "wait". Answer with the JSON object only.

[user] Screenshot attached. Click the target's center or wait.

Decision by decision — 17 decisions

#Bmodel done
B→D
model response (raw)parsedaction (canvas px)D (commit)A (receipt)out tokens
1531 ms1,131 ms
0.60 s
{"action": "wait"}
wait—
——7
21,146 ms1,769 ms
0.62 s
{"action": "wait"}
wait—
——7
31,785 ms2,406 ms
0.62 s
{"action": "wait"}
wait—
——7
42,422 ms3,045 ms
0.62 s
{"action": "wait"}
wait—
——7
53,061 ms3,683 ms
0.62 s
{"action": "wait"}
wait—
——7
63,700 ms4,319 ms
0.62 s
{"action": "wait"}
wait—
——7
74,335 ms4,956 ms
0.62 s
{"action": "wait"}
wait—
——7
84,971 ms5,594 ms
0.62 s
{"action": "wait"}
wait—
——7
95,610 ms6,233 ms
0.62 s
{"action": "wait"}
wait—
——7
106,249 ms6,871 ms
0.62 s
{"action": "wait"}
wait—
——7
116,886 ms7,507 ms
0.62 s
{"action": "wait"}
wait—
——7
127,523 ms9,283 ms
1.76 s
{"action": "click", "x": 874, "y": 374}
click
img (874, 374)
{'kind': 'click', 'x': 279, 'y': 89}
9,283 ms9,303 ms · accepted23
139,330 ms9,933 ms
0.60 s
{"action": "wait"}
wait—
——7
149,949 ms10,571 ms
0.62 s
{"action": "wait"}
wait—
——7
1510,587 ms11,207 ms
0.62 s
{"action": "wait"}
wait—
——7
1611,223 ms11,843 ms
0.62 s
{"action": "wait"}
wait—
——7
1711,859 ms12,481 ms
0.62 s
{"action": "wait"}
wait—
——7

Events and observations

ideventtask timeharness time
1target_appears500 ms500 ms
3effect_visible11,303 ms11,303 ms

797 observations published; first frame depicting the target published at 530 ms. Raw trace: live_d2_none_s409.jsonl.