OpenAI Introduces LifeVal, a Benchmark for Everyday AI Tasks
OpenAI introduced LifeVal on November 2, 2026, an evaluation of how well AI models help with everyday personal-assistant tasks. Its task areas were shaped by a survey of 3,995 respondents, and OpenAI reports unweighted average-reward scores of 49.3% for GPT-5.2 and 80.6% for GPT-6 Astra.
OpenAI introduced LifeVal on November 2, 2026, an evaluation of whether AI models can help people complete practical everyday tasks rather than only answer questions in a chat. Its examples include finding childcare, preparing for school, managing government paperwork, and planning meals. (OpenAI's LifeVal announcement; OpenAI on X)
The evaluation focuses on areas where people say they want help from a personal assistant: caregiving, health, education, household administration, personal finances, shopping, and accessing government services. OpenAI says it used LifeVal to track and drive improvement on everyday tasks over the past year. (OpenAI)
How OpenAI Built LifeVal
OpenAI says it started with the American Time Use Survey and the Multinational Time Use Study, which track how people spend their time in the United States and internationally. The company then added a demographically representative survey of 3,995 respondents to identify the tasks where people most wanted help from a personal assistant. (OpenAI)
The resulting task areas range from caregiving and health to household administration, personal finance, shopping, and government services. That makes LifeVal different from a benchmark limited to knowledge questions: the examples involve work such as comparing childcare options, organizing flood-preparation volunteers, or preparing a care plan. (OpenAI)
Results OpenAI Reports
OpenAI's chart reports an unweighted average reward score for each model release. It lists GPT-5.2 at 49.3%, GPT-5.4 at 73.0%, GPT-5.5 at 77.4%, GPT-5.6 at 80.0%, and GPT-6 Astra at 80.6%. OpenAI reports these results from LifeVal, its evaluation of everyday tasks. (OpenAI)
OpenAI's example cards show a kindergarten search at 93.1%, a flood-preparation volunteer plan at 98.0%, and a meal plan for a parent adjusting to a diabetes diagnosis at 90.4%. (OpenAI)
How to Read the Results
OpenAI says it used LifeVal to track and drive model improvement over the past year, with work on browser use, computer use, and connector tasks. It reports these scores on an evaluation the company says it built. (OpenAI)
For teams evaluating personal assistants, the benchmark points to workflows that go beyond chat responses and into everyday tasks. Its headline chart gives one measure across successive model releases, while the individual examples show specific workflows; neither should be read as a guarantee that another model or a different workflow will receive the same score. (OpenAI)
Read next
Related coverage- OpenAI
OpenAI Launches dots: Always-On Agents in ChatGPT With Their Own Cloud Computer
OpenAI's dots are always-on agents in ChatGPT, powered by GPT-6 Astra, that keep working between conversations using their own cloud computer and the apps you connect. Your first dot is included in Pro and Business Premium at no extra cost, with an allowance for deeper work. Pro access excludes the EEA, Switzerland, and the UK.
- OpenAI
OpenAI Reports Stronger Coding Results in COBOL, Fortran, and Visual Basic
OpenAI says its latest models improved on two software-engineering evaluations for low-resource languages. On MultiLangBench, GPT-6 Astra averaged 87.4% at maximum reasoning effort across 20 programming languages; on a COBOL migration test, GPT-6.1 Sol scored 86.6% at low reasoning effort. These are OpenAI-designed evaluations, not a guarantee of production migration quality.
- OpenAI
OpenAI introduces GeneBench-Pro, a research-level benchmark for agentic computational biology
On June 30, 2026, OpenAI announced GeneBench-Pro: a research-level benchmark meant to measure how well AI agents navigate messy biological data and make the judgment calls real computational biology depends on. OpenAI says GeneBench-Pro contains 129 questions and is open-sourcing 10 representative case studies as a public package on Hugging Face under the MIT License.
- OpenAI
GPT-6.1 Sol: Astra-Level Coding at a Fifth of the Price, With Cost at Every Effort Level
OpenAI released GPT-6.1 Sol on September 29, 2026, at $2 input and $10 output per million tokens, a fifth of GPT-6 Astra's price, with cached input at $0.10. On OpenAI's charts it edges past Astra's best DeepSWE score at high effort and trails Astra by 2.1 points on OSWorld offline. The API model ID is gpt-6.1-sol.
Sources
- OpenAI, LifeVal, November 2, 2026
- OpenAI, LifeVal announcement on X, November 2, 2026
Frequently Asked Questions
What is OpenAI LifeVal?
LifeVal is OpenAI's evaluation of how well AI models can help with practical everyday tasks, including caregiving, health, education, household administration, personal finances, shopping, and government services.
How did OpenAI build LifeVal?
OpenAI says it used two time-use studies and a new demographically representative survey of 3,995 respondents to identify tasks where people most wanted help from a personal assistant.
What scores did OpenAI report?
OpenAI's chart lists unweighted average-reward scores of 49.3% for GPT-5.2, 73.0% for GPT-5.4, 77.4% for GPT-5.5, 80.0% for GPT-5.6, and 80.6% for GPT-6 Astra.