LifeVal: everyday life tasks (LifeVal)
We show this table for reference; we do not rank on it.
OpenAI's benchmark of practical everyday tasks people want a personal assistant to handle, such as finding childcare, accessing government services, budgeting, shopping and caregiving.
Benchmark score on LifeVal — October 29, 2026
We compile the LifeVal rows from provider self-reports. GPT-6 Astra leads the table at 80.6%, followed by GPT-5.6 Sol (80.0%) and GPT-5.5 (77.4%). We do not use these results to rank models overall.
GPT-6 Astra
OpenAI
GPT-5.6 Sol
OpenAI
GPT-5.5
OpenAI
5 modelsAgenticCurrentDisplay onlyUpdated October 29, 2026
Benchmark score table (5 models)
| Rank | Model / configuration | Score | Parameters (B) | Open / closed |
|---|---|---|---|---|
| 1 | GPT-6 AstraOpenAI | 80.6% | Not reported | Closed |
| 2 | GPT-5.6 SolOpenAI | 80.0% | Not reported | Closed |
| 3 | GPT-5.5OpenAI | 77.4% | Not reported | Closed |
| 4 | GPT-5.4OpenAI | 73.0% | Not reported | Closed |
| 5 | GPT-5.2OpenAI | 49.3% | Not reported | Closed |
Among the reported LifeVal rows, GPT-6 Astra is first at 80.6%. The third row is 3.2 points behind. The broader top-10 range is 31.3 points, so the table still separates the published systems.
5 models have been evaluated on LifeVal. The benchmark falls in the Agentic category. LifeVal is currently displayed for reference but excluded from the scoring formula, so it does not directly affect overall rankings.
About LifeVal
Year
2026
Tasks
169 everyday life tasks (caregiving, health, education, household administration, personal finance, shopping, government services)
Format
Average reward, unweighted across tasks
Difficulty
Agentic personal-assistant work
OpenAI built LifeVal from the American Time Use Survey, the Multinational Time Use Study and its own survey of 3,995 people, and reports an unweighted average reward across 169 tasks that use browser use, computer use and connectors. OpenAI ran every published result and has not released the task set, so BenchLM shows the scores display-only and keeps them out of weighted rankings. OpenAI's November 2, 2026 chart names the GPT-5.6 model only as "GPT-5.6"; BenchLM stores it on GPT-5.6 Sol, the flagship tier, as with other unlabelled GPT-5.6 rows. GPT-6 Astra is the launch-blog xhigh setting.
Freshness and provenance
Version
LifeVal (169 tasks)
Refresh cadence
Static
Staleness state
Current
Question availability
Private provider task set with published aggregate results
BenchLM uses freshness metadata to decide whether a benchmark should still be treated as a strong differentiator, a benchmark to watch, or a display-only reference. For the full scoring policy, see the BenchLM methodology page.
Questions
What does LifeVal measure?
OpenAI's benchmark of practical everyday tasks people want a personal assistant to handle, such as finding childcare, accessing government services, budgeting, shopping and caregiving.
Which model scores highest on LifeVal?
GPT-6 Astra by OpenAI currently leads with a score of 80.6% on LifeVal.
How many models are evaluated on LifeVal?
5 AI models have published results on LifeVal in the BenchLM catalog.
Compare top models on LifeVal
Know when it’s worth switching models
The model to choose, the cheaper alternative, and the release we would wait on.
Read a sample issueJoin 6,100+ subscribers.
One email each week. Unsubscribe anytime.