← all runs · ← report

wi007_qwen3_vl_8b_instruct_wi006 / live_d2_harness_pending_s406 not valid

What this trial was

runwi007_qwen3_vl_8b_instruct_wi006triallive_d2_harness_pending_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": "harness_pending", "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 clicks3suppressed proposals3waits10decisions16exact 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 — 16 decisions

#Bmodel done
B→D
model response (raw)parsedaction (canvas px)D (commit)A (receipt)out tokens
1531 ms1,074 ms
0.54 s
{"action": "wait"}
wait—
——7
21,096 ms1,648 ms
0.55 s
{"action": "wait"}
wait—
——7
31,673 ms2,223 ms
0.55 s
{"action": "wait"}
wait—
——7
42,247 ms4,001 ms
1.75 s
{"action": "click", "x": 604, "y": 637}
click
img (604, 637)
{'kind': 'click', 'x': 193, 'y': 152}
4,001 ms4,021 ms · invalid23
54,069 ms4,668 ms
0.60 s
{"action": "wait"}
wait—
——7
64,683 ms5,305 ms
0.62 s
{"action": "wait"}
wait—
——7
75,321 ms5,944 ms
0.62 s
{"action": "wait"}
wait—
——7
85,959 ms7,720 ms
1.76 s
{"action": "click", "x": 604, "y": 654}
suppressed_pending
img (604, 654)
{'kind': 'click', 'x': 193, 'y': 156}
——23
97,763 ms9,508 ms
1.74 s
{"action": "click", "x": 604, "y": 654}
suppressed_pending
img (604, 654)
{'kind': 'click', 'x': 193, 'y': 156}
——23
109,550 ms11,294 ms
1.74 s
{"action": "click", "x": 687, "y": 652}
click
img (687, 652)
{'kind': 'click', 'x': 219, 'y': 156}
11,294 ms11,314 ms · invalid23
1111,338 ms13,082 ms
1.74 s
{"action": "click", "x": 687, "y": 652}
suppressed_pending
img (687, 652)
{'kind': 'click', 'x': 219, 'y': 156}
——23
1213,124 ms14,871 ms
1.75 s
{"action": "click", "x": 607, "y": 652}
click
img (607, 652)
{'kind': 'click', 'x': 194, 'y': 156}
14,871 ms14,891 ms · invalid23
1314,914 ms15,514 ms
0.60 s
{"action": "wait"}
wait—
——7
1415,529 ms16,152 ms
0.62 s
{"action": "wait"}
wait—
——7
1516,168 ms16,790 ms
0.62 s
{"action": "wait"}
wait—
——7
1616,804 ms17,367 ms
0.56 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_harness_pending_s406.jsonl.