New updates and improvements
V7 Go

V7 Go now supports the full GPT‑4.1 family: GPT‑4.1, GPT‑4.1 Mini, and GPT‑4.1 Nano. These models bring stronger reasoning, faster responses, and lower costs across a wide range of tasks.

To keep V7 Go agents running on the best available models, we are also refreshing the full model lineup. The GPT‑4.1 family is available now, with additional models from OpenAI and Google coming soon:

Now live:

• GPT‑4.1 replaces GPT‑4o

• GPT‑4.1 Mini replaces GPT‑4o Mini

• GPT‑4.1 Nano is newly added for high-speed, lightweight tasks

Coming soon:

• OpenAI o3 will replace o1

• OpenAI o4-mini will replace o3-mini

• Gemini 2.5 Pro will replace Gemini 1.5 Pro

Agents powered by GPT‑4.1 are more accurate at following detailed prompts, produce cleaner outputs, and handle longer interactions more reliably. While V7 Go already supports long documents through Knowledge Index and RAG, the expanded 1 million-token context window makes GPT‑4.1 a strong fit for cases where agents need to reference large, complex source documents directly.

Key improvements:

• Agents now understand structured input requirements, edge-case logic, and negative instructions like “don’t include X.”

• GPT-4.1 is 26% cheaper per request on average than GPT-4o, while GPT-4.1 Nano is the fastest, most efficient option for low-latency tasks like classification. Especially relevant for users who use V7 Go with their own API key.

• All 4.1 models can now handle up to 1 million tokens—ideal for agents reviewing large filings, contracts, or due diligence packs without missing context buried deep inside.

V7 Go now offers a conversational approach to AI document analysis through two new features: Cases and Concierge. This update makes interacting with the platform more intuitive and accessible to all users on your team, regardless of their technical background.

You may have noticed that Projects in V7 Go are now called Agents. This change reflects their specialized ability to solve domain-specific tasks using AI through multi-step reasoning, advanced retrieval capabilities, web search, and app integrations. Concierge serves as your AI assistant to communicate with these Agents and delegate tasks to them. Each conversation with Concierge is saved as a new Case and creates a persistent record that you can reference or continue later.

The example above shows Concierge analyzing a CIM document. When a user asks, "Can you triage a CIM document?", the AI assistant identifies and suggests the appropriate Finance CIM Triage agent. After the document is uploaded, the agent extracts key information (such as financial metrics, company description, and background details) and determines whether the triage result is a pass or a fail.

To explore more examples, watch this short tutorial video:

All extracted information is linked to the source via interactive hyperlinks that open the document at the relevant section. This design enables users to verify each data point without manually searching through the document. A collapsible sidebar on the right displays the extracted data alongside its source references, making verification simple and intuitive.

With Cases and Concierge, users can trigger workflows through simple chat messages, use multiple agents in one conversation, attach documents from various sources, ask follow-up questions, and verify AI outputs through interactive references.

The table-style view remains available for setting up agents, automations, and properties. To access it, click the "Configure agent" button that appears when you hover over agents on your V7 Go platform homepage. You’ll also notice that behind the scenes, V7 Go adds a new row each time a specific agent is run via Concierge.

We've created a new, unified panel for managing both your workspace-level files (Library tab) and integrations (Integrations tab), such as Google Drive. This centralized hub allows you to select all files or data sources for use as inputs or references inside your V7 Go projects.

• The Library is now a subsection within Knowledge Hub.

• The Library tab includes new search and sort functionalities for easier file management.

• When editing a prompt, you can inject and mention a file by typing @ or selecting it from the input list

• When selecting an empty File property inside a project, you can choose to:

  • Import a file from your computer

  • Use a file from the Library

  • Connect to Google Drive

At launch, Google Drive is the first available integration. More integrations are coming soon, and we're actively collecting user requests. If there's a specific integration you'd like to see, let us know by filling out this form.

V7 Go now supports Claude 3.7 Sonnet, Anthropic’s latest model with enhanced reasoning and coding capabilities. When selecting models and setting up your projects, you can now choose whether a specific V7 Go property should be calculated using standard Claude 3.7 Sonnet or Claude 3.7 Sonnet with thinking mode enabled for deeper analysis. This addition improves accuracy in instruction-following and multimodal understanding. It also provides greater flexibility and precision to optimize your AI workflows.

Reasoning models, from Deepseek to o3, have taken the world by storm.

Compared to traditional LLMs like GPT4o and Claude Sonnet 3.5, o3 has explicit step-by-step deduction baked into the training process, making it superior for tasks requiring complicated logical reasoning.

o3-mini can now power all property types in V7 Go, including Collections, Numbers, JSON and more.

In addition, we’ve added Gemini Flash 2.0, a powerful workhorse model with low latency and a large (1m) context window.

The Number Property enforces the output of numeric values, making it easier to work with data such as integers, percentages, and currency values.

Common prompts might be:

• Extract the revenue from @company in the @pitch deck. Include only the revenue for the last year. Output in $.

• What percentage of costs in the @CIM come from salaries?

• What is the net income in the @earnings report?

In addition, you can format decimals, negatives, separators, and alignment from the property configuration menu.

Stay tuned, as over the next few days, we’ll be adding both reasoning and AI citations to the Number Property.

Users can now recompute specific property values for individual fields without recalculating an entire property column. If a field produces an error or the output isn’t satisfactory, simply click on the field and select the Recompute option. You can also select a range of values, right-click, and recalculate multiple fields at once.

Until now, you could access and analyze web data in V7 Go by submitting specific URLs. With the new Web Search feature, it is now possible to ask questions directly in your prompt and pull in relevant information from the web.

• V7 Go will automatically use the most effective search queries and visit multiple pages.

• Perfect for data enrichment and accessing current information like stock prices.

• Get results as text, structured JSON, or Select/Multi-select properties.

Note: Source URLs and citations will be automatically displayed in the reasoning tooltip for full transparency in an upcoming update. In the meantime, include "list sources" in your prompt to view reference links in the output.