← all runs · ← report

wi007_qwen3_vl_8b_instruct_wi006 / live_d8_none_s408 valid

What this trial was

runwi007_qwen3_vl_8b_instruct_wi006triallive_d8_none_s408conditionlivescenariopending_effectskinnotificationcanvas320×240 px (target 24 px)event at500 msresponse window16 sframe period16.67 msobservation publication delay30 msdeliverysimulated_fixed_delay, 20 msclockmonotonic_nseffect_delay_ns8000000000irrelevant_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→A11.56 sduplicates0invalid clicks3suppressed proposals0waits18decisions22exact successTruedropped0contract violationsnone

Replay

animated replay

Recorded frames from the first decision to the last receipt, every 4th frame, time compressed 10×; red = committed click.

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 — 22 decisions

#Bmodel done
B→D
model response (raw)parsedaction (canvas px)D (commit)A (receipt)out tokens
1531 ms1,063 ms
0.53 s
{"action": "wait"}
wait—
——7
21,077 ms1,638 ms
0.56 s
{"action": "wait"}
wait—
——7
31,652 ms2,213 ms
0.56 s
{"action": "wait"}
wait—
——7
42,227 ms2,787 ms
0.56 s
{"action": "wait"}
wait—
——7
52,802 ms3,362 ms
0.56 s
{"action": "wait"}
wait—
——7
63,376 ms4,000 ms
0.62 s
{"action": "wait"}
wait—
——7
74,016 ms4,637 ms
0.62 s
{"action": "wait"}
wait—
——7
84,652 ms5,275 ms
0.62 s
{"action": "wait"}
wait—
——7
95,291 ms5,850 ms
0.56 s
{"action": "wait"}
wait—
——7
105,865 ms6,485 ms
0.62 s
{"action": "wait"}
wait—
——7
116,500 ms7,123 ms
0.62 s
{"action": "wait"}
wait—
——7
127,139 ms7,763 ms
0.62 s
{"action": "wait"}
wait—
——7
137,779 ms8,402 ms
0.62 s
{"action": "wait"}
wait—
——7
148,418 ms9,038 ms
0.62 s
{"action": "wait"}
wait—
——7
159,053 ms9,674 ms
0.62 s
{"action": "wait"}
wait—
——7
169,690 ms10,260 ms
0.57 s
{"action": "wait"}
wait—
——7
1710,274 ms12,039 ms
1.76 s
{"action": "click", "x": 805, "y": 505}
click
img (805, 505)
{'kind': 'click', 'x': 257, 'y': 121}
12,039 ms12,059 ms · accepted23
1812,087 ms12,621 ms
0.53 s
{"action": "wait"}
wait—
——7
1912,635 ms14,335 ms
1.70 s
{"action": "click", "x": 687, "y": 487}
click
img (687, 487)
{'kind': 'click', 'x': 219, 'y': 116}
14,335 ms14,355 ms · invalid23
2014,377 ms16,051 ms
1.67 s
{"action": "click", "x": 604, "y": 477}
click
img (604, 477)
{'kind': 'click', 'x': 193, 'y': 114}
16,051 ms16,071 ms · invalid23
2116,093 ms17,768 ms
1.67 s
{"action": "click", "x": 687, "y": 484}
click
img (687, 484)
{'kind': 'click', 'x': 219, 'y': 116}
17,768 ms17,788 ms · invalid23
2217,788 ms18,343 ms
0.55 s
{"action": "wait"}
wait—
——7

Events and observations

ideventtask timeharness time
1target_appears500 ms500 ms
2window_closes16,500 ms16,500 ms

1020 observations published; first frame depicting the target published at 530 ms. Raw trace: live_d8_none_s408.jsonl.