← All posts

3 min read GenViz Team conversation analyticsAI modelsrelease

GPT 6.1 Sol joins the GenViz stable

GPT 6.1 Sol is now in GenViz. Explore business data with natural-language questions, inspect SQL results, and choose reasoning effort for conversation analytics.

GPT 6.1 Sol has joined the GenViz stable. Choose it in the OpenAI section of the model picker to explore your business data through conversation: ask a question, inspect the results, and follow up with the next thing you want to understand.

It joins GPT 6 Sol and Luna and GPT 6 Astra, using the same database connections, read-only query tools, and analysis results already available in GenViz.

GPT 6.1 Sol for conversation analytics

In its GPT 6.1 Sol announcement, OpenAI positions the model as offering performance close to Astra at a lower cost for complex work. For business data analysis, a useful place to try it is a question that takes several steps: establish a total, break it down, and investigate what contributed to it.

In GenViz, those steps stay together in a conversation. You can review a SQL query, open its result table, and ask for a chart or a further calculation. When you need to work across saved results, Python analysis gives you another way to examine the data and review the code behind a calculation.

From an overdue invoice total to the customers behind it

Start with a question that has a clear business definition:

Show outstanding invoice balances as of today, grouped into 1–30, 31–60, 61–90, and more than 90 days overdue. Exclude voided invoices and show the totals in a table.

Open the table and inspect the query. Check that it uses the due date, accounts for partial payments, and treats credits the way your team expects. Then ask a follow-up:

Which five customers account for the largest outstanding balance more than 60 days overdue? Show their balances and share of that total.

Ask for a chart to compare those customers, or open the underlying invoice rows to understand a particular balance. Each question builds on results you can check, with saved extracts available for later inspection.

Adjust reasoning effort for your question

GPT 6.1 Sol starts at medium reasoning effort in GenViz. The model picker offers low, medium, high, xhigh, and max. A focused lookup and an investigation involving several joins may call for different settings; start with medium and compare using questions you can verify.

GPT 6.1 Sol always uses reasoning, so its picker does not include a none option. The setting changes the reasoning effort available to the model while keeping your GenViz analysis tools at hand.

Try GPT 6.1 Sol in GenViz

Open the model picker and choose OpenAI → GPT 6.1 Sol for your next question. Availability follows your organization’s OpenAI access. On the Team plan, an administrator needs to configure an OpenAI API key with access to the model.

Your prompts and the context used to answer them, including relevant schema information and query results, are sent to OpenAI. Our privacy policy explains the services involved.

Bring a question from your next business review and work through it one result at a time.

Explore conversation analytics or download GenViz to get started.