ChainWorld
Agents complete sequences of desktop tasks while preserving the state needed for later steps. OSWorld checks each step.
The authors tested four agents on 347 sequences. They presented the tasks together or revealed them one at a time.
- Made by
- Scale AI and UC Berkeley
- Released
- June 2026
- Kind
- Desktop apps
- Level
- Everyday
- Scored
- Checked automatically
- Answers
- Agent task
- Measures
- Capability
- Language
- English
- Interaction
- Agentic
- Environment
- Desktop apps
- Items from
- Another benchmark, Generated
- Refresh
- One fixed set
- Run by
- Anyone
- Based on
- OSWorld
- In views
- Computer use, Agents
- Strongest model tested
- GPT-5.5 #21 on the capability index
- Last model added
- April 2026
- Scores refreshed
- November 20th, 2026
GPT-5.5 leads with 31% among models released by April 2026
Best score per model from the benchmark's paper, read on November 20th, 2026. A paper’s table doesn’t add newer models. Where a model ran with several settings or harnesses, we keep its best. Every run is in the next chart.
The newest model with a score here came out in April 2026, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:
- GPT-6.2 Astra
- Claude Fable 5.2
-
Gemini 4 Carbon
- GPT-6 Astra
- and 16 more
- 1 GPT-5.5 Multi-turn 31%
- 2 Claude Opus 4.7 Multi-turn 17%
- 3 Kimi K2.5 Multi-turn 10%
- 4
Gemini 3.1 Pro Single turn 3.00%
Source: The ChainWorld paper
| # | Model | Score ± | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | GPT-5.5 | 31% ±5 | Multi-turn | Apr 2026 | Official leaderboard |
| 2 | Claude Opus 4.7 | 17% ±4 | Multi-turn | Apr 2026 | Official leaderboard |
| 3 | Kimi K2.5 | 10% ±3 | Multi-turn | Jan 2026 | Official leaderboard |
| 4 | | 3.00% ±2 | Single turn | Feb 2026 | Official leaderboard |
Scores at each effort level
Models the leaderboard ran at two or more effort levels. We don’t have enough cost data here to say what the extra effort costs.
Source: The ChainWorld paper
| Model | Effort | Harness | Score | Cost per task | Output tokens |
|---|---|---|---|---|---|
| GPT-5.5 | Multi-turn | 31% | |||
| Claude Opus 4.7 | Multi-turn | 17% | |||
| GPT-5.5 | Single turn | 15% | |||
| Claude Opus 4.7 | Single turn | 11% | |||
| Kimi K2.5 | Multi-turn | 10% | |||
| Kimi K2.5 | Single turn | 5.00% | |||
| Single turn | 3.00% | ||||
| Multi-turn | 2.00% |
Number one over time
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: The ChainWorld paper
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: Official ChainWorld 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
- 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
7 of 9 checks pass, 2 warnings. Look first at the Task access check. We don’t know whether the tasks are public. Every benchmark
Catalog record 1 warning 1 pass
- Warning Task access We don’t know whether the tasks are public.
- Pass Confidence Checked against a primary source.
Scores 1 warning 2 pass 2 notes
- Pass Independent runs Scores from the benchmark’s paper (4 entries).
- Note Models scored Only 4 models we track.
- Warning Newest score The newest came 7 months ago, in April 2026, for GPT-5.5.
- Note Frontier models None of the top 10 has a score. The strongest tested is GPT-5.5 (#21).
- Pass Headroom The best is 31%, under 85%, so it still separates models.
Leaderboard import 3 pass
- Pass Importer We read the board on arxiv.org: 8 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.