LLM Changelog

Collect user feedback in Copilot Studio agent builder
[Web]
Makers can also now submit feedback-compliments, problems, or suggestions-directly to the product team from inside the Microsoft 365 agent builder. This can be done at any stage of the authoring process, including comments and optional metadata for troubleshooting. This feature allows quick issue reporting without disrupting workflow and enables the product team to address feedback efficiently.
Learn more
.

Our 2.5 Pro (experimental) model is now available to all Gemini users, with CanvasWhat:Today we’re expanding access to our most intelligent AI model, 2.5 Pro (experimental), to all Gemini users. This state-of-the-art model has thinking capabilities natively built in, with exceptional performance in coding, math, image understanding and more.Canvas, our new interactive space that makes it easy to create, refine, and share your work, is now available to try with 2.5 Pro (experimental). With 2.5 Pro’s improved coding capabilities, coupled with Canvas, you can quickly create compelling web apps or generate code for immediate use.Gemini users will be able to try 2.5 Pro (experimental) with rate limits, and Gemini Advanced users will continue to have expanded access and a significantly larger context window.Being an experimental model, it can have unexpected behaviors and may make mistakes.Why:We want to bring the best model in the world to all Gemini users. Your feedback helps us improve these models over time and learning from experimental launches informs how we release models more widely.

Our 2.5 Pro (experimental) model is now available to all Gemini users, with CanvasWhat:Today, we're expanding access to our most intelligent AI model, 2.5 Pro (experimental), to all Gemini users. This state-of-the-art model has thinking capabilities natively built in, with exceptional performance in coding, maths, image understanding and more.Canvas, our new interactive space that makes it easy to create, refine and share your work, is now available to try with 2.5 Pro (experimental). With 2.5 Pro's improved coding capabilities, coupled with Canvas, you can quickly create compelling web apps or generate code for immediate use.Gemini users will be able to try 2.5 Pro (experimental) with rate limits, and Gemini Advanced users will continue to have expanded access and a significantly larger context window.Being an experimental model, 2.5 Pro may behave unexpectedly or make mistakes.Why: We want to bring the best model in the world to all Gemini users. Your feedback helps us improve these models over time, and learning from experimental launches informs how we release models more widely.

We’ve made improvements to GPT-4o—it now feels more intuitive, creative, and collaborative, with enhanced instruction-following, smarter coding capabilities, and a clearer communication style.
Smarter problem-solving in STEM and coding:
GPT-4o has further improved its capability to tackle complex technical and coding problems. It now generates cleaner, simpler frontend code, more accurately thinks through existing code to identify necessary changes, and consistently produces coding outputs that successfully compile and run, streamlining your coding workflows.
Enhanced instruction-following and formatting accuracy:
GPT-4o is now more adept at following detailed instructions, especially for prompts containing multiple or complex requests. It improves on generating outputs according to the format requested and achieves higher accuracy in classification tasks.
“Fuzzy” improvements:
Early testers say that the model seems to better understand the implied intent behind their prompts, especially when it comes to creative and collaborative tasks. It’s also slightly more concise and clear, using fewer markdown hierarchies and emojis for responses that are easier to read, less cluttered, and more focused. We're curious to see if our users also find this to be the case.
This model is now available in ChatGPT and in the API as the newest snapshot of chatgpt-4o-latest. We plan to bring these improvements to a dated model in the API in the coming weeks.

Priority access with Gemini Advanced: Try our latest 2.5 Pro (experimental) modelWhat:Today we’re introducing Gemini 2.5, our most intelligent AI model. Our first 2.5 release is a chat optimized version of Gemini-2.5-Pro-Exp-03-25, which is state-of-the-art on a wide range of benchmarks and debuts at #1 on LMArena by a significant margin. This model also has thinking capabilities natively built in, with improved performance in complex tasks like coding, math, and image understanding.2.5 Pro (experimental) is now rolling out to the Gemini web and mobile app and is available to qualifying Google Workspace business and education plans.This experimental model is meant to be an early preview and can have unexpected behaviors and may make mistakes.Why:We believe in rapid iteration and bringing the best of Gemini to the world, and we want to give Gemini Advanced subscribers priority access to our latest AI innovations. Your feedback helps us improve these models over time and learning from experimental launches informs how we release models more widely.

Priority access with Gemini Advanced: Try our latest 2.5 Pro (experimental) modelWhat:Today, we're introducing Gemini 2.5, our most intelligent AI model. Our first 2.5 release is a chat-optimised version of Gemini-2.5-Pro-Exp-03-25, which is state of the art on a wide range of benchmarks and debuts at #1 on LMArena by a significant margin. This model also has thinking capabilities natively built in, with improved performance in complex tasks like coding, maths and image understanding.2.5 Pro (experimental) is now rolling out to the Gemini web and mobile app and is available to qualifying Google Workspace business and education plans.This experimental model is meant to be an early preview; it may behave unexpectedly or make mistakes.Why: We believe in rapid iteration and bringing the best of Gemini to the world, and we want to give Gemini Advanced subscribers priority access to our latest AI innovations. Your feedback helps us improve these models over time, and learning from experimental launches informs how we release models more widely.

Enhanced transparency for declarative agent metadata
Admins can now view granular metadata in the Microsoft 365 admin center—including data source details and custom actions—for declarative agents. This clarity enables more informed decisions on app availability and management across your organization.

Copilot on Edge Update
[Web]
Copilot on Edge has been updated and requires users to update their version of Edge to version 134.0.3124.51 or newer to receive the latest functionality. This update includes file upload availability on the web tab and a smoother authentication experience via the Edge Work Profile.

Enhanced transparency for declarative agent metadata
Admins can now view granular metadata in the Microsoft 365 admin center-including data source details and custom actions-for declarative agents. This clarity enables more informed decisions on app availability and management across your organization.

Try Canvas, a new way to collaborate with GeminiWhat:Starting today, you can collaborate with Gemini 2.0 Flash to write documents and code in Canvas, a new interactive space that makes refining & sharing your work really easy.Co-create documents: Generate a first draft, then rapidly refine and ask Gemini for feedback on your edits. Update specific sections or the whole draft, and use the quick editor tools to change the tone, length, or formatting. From essays to blog posts to reports, uplevel your documents in Canvas.Generate & iterate on code: Easily convert your ideas to working prototypes for web apps, Python scripts, and more. Ask Gemini to generate & preview React or HTML code directly in Canvas in a familiar code editor, and review Gemini's changes on each turn.Canvas is available globally across all Gemini-supported languages. Select Canvas in the prompt bar to start creating!Why:Canvas lets you experience the power of collaboration with Gemini because you can see your ideas and iterations take shape in real time. Go from blank slate to a share-worthy creation in minutes. This means you can focus on your vision to create something awesome and leave the heavy lifting of generating, editing, and fixing things to Gemini.Introducing Audio Overview, now in GeminiWhat:Leveraging the same technology that powers NotebookLM’s Audio Overviews, Gemini users can now generate podcast-style conversations based on documents, slides, and Deep Research reports. Upload files about topics you want to explore and enjoy dynamic discussions between two AI hosts with unique perspectives.Why:Stay informed while you’re on the go and gain new insights by experiencing your content in a fresh way.

Try Canvas, a new way to collaborate with GeminiWhat:Starting today, you can collaborate with Gemini 2.0 Flash to write documents and code in Canvas, a new interactive space that makes refining & sharing your work really easy.Co-create documents: Generate a first draft, then rapidly refine and ask Gemini for feedback on your edits. Update specific sections or the whole draft, and use the quick editor tools to change the tone, length or formatting. From essays to blog posts to reports, uplevel your documents in Canvas.Generate & iterate on code: Easily convert your ideas to working prototypes for web apps, Python scripts, and more. Ask Gemini to generate & preview React or HTML code directly in Canvas in a familiar code editor, and review Gemini's changes on each turn.Canvas is available globally across all Gemini-supported languages. Select Canvas in the prompt bar to start creating!Why: Canvas lets you experience the power of collaboration with Gemini because you can see your ideas and iterations take shape in real time. Go from blank slate to a share-worthy creation in minutes. This means you can focus on your vision to create something awesome and leave the heavy lifting of generating, editing, and fixing things to Gemini.Introducing Audio Overview, now in GeminiWhat:Leveraging the same technology that powers NotebookLM’s Audio Overviews, Gemini users can now generate podcast-style conversations based on documents, slides, and Deep Research reports. Upload files about topics that you want to explore and enjoy dynamic discussions between two AI hosts with unique perspectives.Why: Stay informed while you're on the go and gain new insights by experiencing your content in a fresh way.

1. In-line message error retries: If you encounter a message error, you can retry or just continue chatting in the same conversation.
2. o1 and o3-mini now offer Python-powered data analysis in ChatGPT: You can now ask these models to perform tasks like running regressions on test data, visualizing complex business metrics, and conducting scenario-based simulations.
3. Conversation Drafts: Unsubmitted messages in your message prompt will now be saved.
4. New UI for Temporary Chats: We’ve made accessing and viewing these private conversations more clear.

Update to Gemini 2.0 Flash Thinking (experimental) in GeminiWhat:Starting today, an improved version of Gemini 2.0 Flash Thinking (experimental) will become available to Gemini app users. Built on the foundation of 2.0 Flash, this model delivers improved performance and better advanced reasoning capabilities with efficiency and speed.Starting in English, 2.0 Flash Thinking (experimental) now works with your favorite Gemini features and connected apps such as YouTube, Maps, Search and more. Gemini Advanced users will also have access to a 1M token context window with this model.Why:We're investing in thinking and reasoning capabilities because we believe they unlock deeper intelligence and deliver enhanced performance for tasks requiring complex reasoning, such as coding, scientific discovery, and advanced math.Deep Research is now powered by Gemini 2.0 Flash Thinking (experimental) and available to more Gemini users at no costWhat:Today, we're upgrading Gemini Deep Research to use the 2.0 Flash Thinking (experimental) model. We're also giving Gemini users the ability to try Deep Research at no cost. Gemini Advanced and qualifying enterprise users have expanded access to Deep Research to save even more time on their most complex projects.With Deep Research, you can save hours of work as Gemini analyzes relevant information on your behalf to create comprehensive multi-page reports on any topic in minutes. With advanced reasoning from 2.0 Flash Thinking, Gemini is even better at all stages of research from planning to searching to reporting.Why:By powering Deep Research with this upgraded model, and expanding its availability, Deep Research is taking one step forward in Google's mission to organize the world's information and make it universally accessible and useful.Gemini gets personal, with tailored help based on your Search historyWhat:Starting today, Gemini can nowuse your Search historyto give you even more helpful and personalized responses. When you select "Per

Update to Gemini 2.0 Flash Thinking (experimental) in GeminiWhat: Starting today, an improved version of Gemini 2.0 Flash Thinking (experimental) will become available to Gemini app users. Built on the foundation of 2.0 Flash, this model delivers improved performance and better advanced reasoning capabilities with efficiency and speed.Starting in English, 2.0 Flash Thinking (experimental) now works with your favourite Gemini features and connected apps such as YouTube, Maps, Search and more. Gemini Advanced users will also have access to a 1M token context window with this model.Why:We're investing in thinking and reasoning capabilities because we believe they unlock deeper intelligence and deliver enhanced performance for tasks requiring complex reasoning, such as coding, scientific discovery, and advanced math.Deep Research is now powered by Gemini 2.0 Flash Thinking (experimental) and available to more Gemini users at no costWhat:Today, we're upgrading Gemini Deep Research to use the 2.0 Flash Thinking (experimental) model. We're also giving Gemini users the ability to try Deep Research at no cost. Gemini Advanced and qualifying enterprise users have expanded access to Deep Research to save even more time on their most complex projects.With Deep Research, you can save hours of work as Gemini analyses relevant information on your behalf to create comprehensive multi-page reports on any topic in minutes. With advanced reasoning from 2.0 Flash Thinking, Gemini is even better at all stages of research from planning to searching to reporting.Why:By powering Deep Research with this upgraded model, and expanding its availability, Deep Research is taking one step forward in Google's mission to organize the world's information and make it universally accessible and useful.Gemini gets personal, with tailored help based on your Search historyWhat:Starting today, Gemini can nowuse your Search historyto give you even more helpful and personalized responses. When you select 'P

**Model updates:**

- Launched an experimental [Gemini 2.0 Flash](https://ai.google.dev/gemini-api/docs/image-generation#gemini) model capable of image generation and editing.
- Released `gemma-3-27b-it`, available on [AI Studio](https://aistudio.google.com) and through the Gemini API, as part of the [Gemma 3](https://ai.google.dev/gemma/docs/core) launch.

**API updates:**

- Added support for [YouTube URLs](https://ai.google.dev/gemini-api/docs/vision#youtube) as a media source.
- Added support for including an [inline video](https://ai.google.dev/gemini-api/docs/vision#inline-video) of less than 20MB.

Enrichment of Messages in Copilot Chat
[Web]
This feature enhances your communication experience by making it easier to understand and interact with your chat messages in Copilot Chat. With this feature, you will see cards and hoverable experiences that provide additional details of the chat without leaving your current view. This means you can quickly grasp the context of your conversations and find the information you need more efficiently.