← all runs · ← report

wi007_qwen3_vl_8b_instruct_wi006 / live_d2_none_s406 not valid

What this trial was

runwi007_qwen3_vl_8b_instruct_wi006triallive_d2_none_s406conditionlivescenariopending_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 acceptedFalsefirst E→A—duplicates0invalid clicks6suppressed proposals0waits12decisions18exact successFalsedropped0contract 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 — 18 decisions

#Bmodel done
B→D
model response (raw)parsedaction (canvas px)D (commit)A (receipt)out tokens
1531 ms1,128 ms
0.60 s
{"action": "wait"}
wait—
——7
21,143 ms1,766 ms
0.62 s
{"action": "wait"}
wait—
——7
31,783 ms2,402 ms
0.62 s
{"action": "wait"}
wait—
——7
42,430 ms3,043 ms
0.61 s
{"action": "wait"}
wait—
——7
53,071 ms3,682 ms
0.61 s
{"action": "wait"}
wait—
——7
63,711 ms4,321 ms
0.61 s
{"action": "wait"}
wait—
——7
74,348 ms4,962 ms
0.61 s
{"action": "wait"}
wait—
——7
84,990 ms5,600 ms
0.61 s
{"action": "wait"}
wait—
——7
95,629 ms6,239 ms
0.61 s
{"action": "wait"}
wait—
——7
106,268 ms8,018 ms
1.75 s
{"action": "click", "x": 604, "y": 654}
click
img (604, 654)
{'kind': 'click', 'x': 193, 'y': 156}
8,018 ms8,038 ms · invalid23
118,086 ms9,829 ms
1.74 s
{"action": "click", "x": 602, "y": 652}
click
img (602, 652)
{'kind': 'click', 'x': 192, 'y': 156}
9,829 ms9,849 ms · invalid23
129,874 ms11,618 ms
1.74 s
{"action": "click", "x": 607, "y": 652}
click
img (607, 652)
{'kind': 'click', 'x': 194, 'y': 156}
11,618 ms11,638 ms · invalid23
1311,662 ms13,405 ms
1.74 s
{"action": "click", "x": 607, "y": 652}
click
img (607, 652)
{'kind': 'click', 'x': 194, 'y': 156}
13,405 ms13,425 ms · invalid23
1413,449 ms15,190 ms
1.74 s
{"action": "click", "x": 604, "y": 654}
click
img (604, 654)
{'kind': 'click', 'x': 193, 'y': 156}
15,190 ms15,210 ms · invalid23
1515,235 ms15,836 ms
0.60 s
{"action": "wait"}
wait—
——7
1615,852 ms16,474 ms
0.62 s
{"action": "wait"}
wait—
——7
1716,490 ms18,254 ms
1.76 s
{"action": "click", "x": 602, "y": 645}
click
img (602, 645)
{'kind': 'click', 'x': 192, 'y': 154}
18,254 ms18,274 ms · invalid23
1818,274 ms18,827 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_d2_none_s406.jsonl.