Cody is built on model interoperability and we aim to provide access to the best and latest models, and today we're making an update to the default models offered to Enterprise customers.

DeepSeek-V2 for autocomplete

In our quest for consistency across how you use Cody, JetBrains users now get the same side panel experience found in Cody for VS Code and on the web. This update includes improvements such as:

• Chat lives directly in the side panel: Chatting with Cody happens primarily in the side panel, with the option to open a dedicated chat window.
• Dedicated tabs in the side panel: Toggle between chat, history, Prompt Library, setting, and account at the top of the sidebar, making it easier to navigate between commonly performed actions.
• Prompts are easier to discover and use: The new Prompts and Prompt Library appear underneath the chat, making it easier to use and create Prompts while chatting with Cody.

Announced for VS Code users in the last IDE release (v1.32), Smart Apply is now available for JetBrains and allows users to take suggestions from the Cody chat window and near-instantly turn them into diffs in your code. Every code block presented in a chat window will now have a new "Apply" button shown to the user and when clicked, Cody will automatically analyze your open code file, find where that relevant code should live, and add a diff.

The most common use case when searching your codebase is to find the best match to your query as quickly as possible. However, for some use cases such as audits and security checks, completeness of results are far more important. Search Jobs allow search queries to be run across your entire codebase (including repos, branches, and revisions) at scale and prioritize completeness over quick response times and results ranking.

In beta and available by default since Sourcegraph 5.3, we're happy to announce that Search Jobs are now generally available. Learn more about Search Jobs in our docs.

You can now control which parts of a multi-line edit to accept with more granularity. When you ask Cody to edit a block of code, the presented diff will be split into discrete code blocks, and you can accept or reject each diff separately.

Huge thank you to Justin Milner for making this contribution!

We heard from users that the chat wasn't using the full sidebar width, and the chat text was narrower than needed. We've updated this to make chat more space-efficient.

Autocomplete has been overhauled with multiple under-the-hood changes, making it both faster and more accurate:

• P75 latency is down by 350 milliseconds
• Multi-line completions are significantly more accurate and relevant to the surrounding context
• The average completion acceptance rate (CAR) for early users is up by >4%

The Ask Cody to Fix button (released in v6.0.25) is now available on Windows. With just a single button, you can go from seeing an error in your IDE to implementing a fix.

Chat-oriented programming (CHOP) is a new paradigm for chatting with AI to work through problems, write code, and build solutions. CHOP is now even more powerful with the Smart Apply feature, which lets you take suggestions from Cody in the chat and near-instantly turn them into diffs in your code.

Whenever Cody provides a code block as a suggestion in chat, press Apply, and Cody will analyze your open code file, find where that relevant code should live, and introduce a diff. For chat messages where Cody presents multiple code suggestions, you can apply each in sequence to go from chat suggestions to written code in only a few seconds.