Skip to content
All AndroidWorld tasks

Audio Recorder Record Audio With File Name

Record an audio clip and save it with name “2023_06_24_training.m4a” using Audio Recorder app.

Task outcome
Pass (Google’s grader: fail; it reads the instruction differently, see the † note)
Pass on Mobilerun’s grader, fail on Google’s1
Time per task
42.3 s
vs. Mobilerun 250.9 s25.9× faster
AI model cost
$0.0132
vs. Mobilerun ~$0.773~58× cheaper

Recording and steps

0requestInput

No memory vs “shared memory”

Mobilerun used “shared memory” carried lessons from task to task. Lusor agent ran with clean context and no memory of prior task runs or attempts.

No self-improvement loop vs “self-improved”

Mobilerun’s self-improvement agent reviewed its own benchmark runs and rewrote its app cards and memory before each evaluation round. Lusor’s agent doesn’t learn between runs: our team wrote its six app skills and general guidelines by hand, and all of them are published.

Select skills vs “app cards”

Mobilerun used per-app instruction files (“app cards”); which apps had them, and what they said, wasn’t published. Lusor created small skills for 6 of the 21 AndroidWorld apps. The agent had to decide to use them. They are fully published.

Run details

Recorded
November 9, 2026
Phone
Android 13 emulator, reset to the benchmark snapshot
Lusor tool calls
16

† The app adds .m4a itself, so Lusor typed 2023_06_24_training and saved 2023_06_24_training.m4a. Google’s grader looks for the file you get by typing the full name (2023_06_24_training.m4a.m4a). Mobilerun’s fork checks for the name as written. 1Grader: Mobilerun’s fork of the AndroidWorld grader (c5467e6); Google’s unmodified grader (30583e2) also shown. 2Speed = Mobilerun’s published session time ÷ ours. 3Mobilerun est. $ = their published action count × $0.0766 (one GPT-6 Astra low call per action at the per-call cost we measured); their token use is not published, so it is an estimate, not their bill. “Skill assisted” = the agent read an app skill that has rules for that task (research’s reading of each skill’s text); reading a skill that does not apply is not marked.