Benchmark Almanac

PhoneShell iOS Agent Benchmark

Kimi K3 96.6% leads since September 2026
Not in model cards

Agents operate apps on a physical iPhone to set timers, find books, and complete other everyday tasks. Automated checks read the resulting device state.

Blolabel compares six models on the same 58 tasks. No pair differed significantly. The full suite includes 80 public tasks and 16 withheld tasks.

Made by
Blolabel
Released
September 2026
Kind
Phones
Level
Everyday
Scored
Checked automatically
Answers
Agent task
Tasks
Public, plus a private test set
Measures
Capability
Language
English
In and out
Text, Screens → Actions
Interaction
Agentic
Refresh
One fixed set
Run by
A host
Strongest model tested
Claude Opus 4.8 #22 on the capability index
Last model added
July 2026
Scores refreshed
November 20th, 2026
6
Models scored on its own leaderboard

Kimi K3 leads with 96.6% among models released by July 2026

Best score per model from the benchmark's own leaderboard, read on November 20th, 2026. Where a model ran with several settings or harnesses, we keep its best.

The newest model with a score here came out in July 2026, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:

  1. 1 Kimi K3 96.6%
  2. 2 DeepSeek-V3.2 93.1%
  3. 3 Gemini 3.1 Pro 93.1%
  4. 4 GPT-5.1 91.4%
  5. 5 Claude Opus 4.8 91.4%
  6. 6 Qwen3-Max 89.7%

Source: Official PhoneShell iOS Agent Benchmark leaderboard

#ModelScore ±SettingDateSource
1 Kimi K3 96.6% 88 to 99 Jul 2026 Official leaderboard
2 DeepSeek-V3.2 93.1% 84 to 97 Dec 2025 Official leaderboard
3 Gemini 3.1 Pro 93.1% 84 to 97 Feb 2026 Official leaderboard
4 GPT-5.1 91.4% 81 to 96 Nov 2025 Official leaderboard
5 Claude Opus 4.8 91.4% 81 to 96 May 2026 Official leaderboard
6 Qwen3-Max 89.7% 79 to 95 Sep 2025 Official leaderboard

Number one over time

Lab claims count from the day the card came out, independent runs from the model’s release.

Moonshot AI
Own leaderboard Passed 85% in Sep 2026, by Kimi K3
Kimi K3 96.6%

Source: Official PhoneShell iOS Agent Benchmark leaderboard

Qwen3-Max on the official leaderboard passed 85% in September 2025

Models released before it came out already scored that high. All the scores we have, one dot each, so a model tested at several settings appears more than once.

Plot with other benchmarks
On the benchmark’s own leaderboard Best so far
0% 20% 40% 60% 80% 100% 2026 Near the ceiling (85%) Kimi K3 DeepSeek-V3.2 0% 20% 40% 60% 80% 100% 2026 Near the ceiling (85%) Kimi K3 GPT-5.1

Source: Official PhoneShell iOS Agent Benchmark leaderboard

No lab model card reports it yet

We read the headline results of 540 launch posts from 57 labs, and none of them list this benchmark.

More in Computer and browser use

All 119 in Computer and browser use
  • BrowseComp In use Hard-to-find facts that need persistent web browsing. Short verifiable answers graded for exact correctness. Claude Fable 5 93.3%
  • ScreenSpot-Pro In use GUI grounding on high-resolution professional-software screenshots. The model must locate the target UI element. GPT-6 Astra 92.7%
  • AndroidWorld In use Agents complete 116 tasks across 20 real Android apps in an emulator. Each task checks the device's system state to judge success. GPT-6 Astra 100%
  • OSWorld 2.1 In use Agents complete 108 long-horizon workflows across 31 self-hosted websites and desktop apps, averaging 318 tool calls. Scored by binary completion and checkpoint partial credit. Claude Opus 5 44.3%
  • WebArena-Verified In use Agents complete 812 audited WebArena tasks on self-hosted websites. Deterministic evaluators check the agent's answer and captured network traces. Muse Spark 1.1 69%
  • OSWorld 2.1 (offline subset) In use Agents complete the 82 OSWorld 2.1 tasks on the official offline list, with no internet access. Scored by checkpoint partial credit and strict pass rate. Claude Fable 5.2 89.3%

Data audit

All 9 checks pass. Every benchmark

Catalog record 2 pass
  • Pass Task access Public tasks, plus a private test set.
  • Pass Confidence Checked against a primary source.
Scores 3 pass 2 notes
  • Pass Independent runs Scores from the official leaderboard (6 entries).
  • Pass Models scored 6 models we track.
  • Pass Newest score The newest is from July 2026, for Kimi K3.
  • Note Frontier models None of the top 10 has a score, which is fine for a retired test. The strongest tested is Claude Opus 4.8 (#22).
  • Note Headroom Saturated. The best is 96.6%, past 85% since September 2025.
Leaderboard import 3 pass
  • Pass Importer We read the board on blolabel.ai: 6 rows for models we track at the last read.
  • Pass Last read November 20th, 2026.
  • Pass Read worked The last read worked.
Consistency 1 pass 1 note
  • Note Lab claims No model has both a lab number and an independent run.
  • Pass Known issues No published audit found problems.

How we check