Google DeepMind’s music-generation model, Lyria, creates high-fidelity instrumental tracks directly from your text prompts. With detailed control over style, instruments, and tempo, Lyria empowers you to transform your creative ideas into music seamlessly through chat.

As the first music generation model available on Poe, Lyria also allows users to build custom experiences with Poe Apps, unlocking even more possibilities for creativity.

Try Lyria now at https://poe.com/Lyria and enjoy its availability across all platforms. We’re excited to see the incredible music you’ll create!

FLUX Kontext, the latest FLUX image model, delivers advanced image generation and conversational image editing. Its multi-turn editing lets you refine images quickly with text prompts using our “Remix” button - allowing you to preserve character consistency while editing and even restyling your images. 

Two Powerful Versions:

• FLUX Kontext Pro: Perfect for fast, step-by-step editing while maintaining consistent people and styles.

• FLUX Kontext Max: Delivers sharper details and superior text rendering for high-quality creations.

You can try FLUX Kontext Pro at https://poe.com/Flux-Kontext-Pro, Max at https://poe.com/Flux-Kontext-Max, and across all platforms. 

Gemini 2.0 Flash supports native image output and conversational editing, allowing you to create and refine images by describing what you want.

Unique to this bot, you can also generate text and images together in one request. For example, it can create a short story about a topic while generating images for each scene based on your direction.

You can try native image generation on Gemini 2.0 Flash at https://poe.com/Gemini-2.0-Flash-Preview and on all platforms where Poe is available.

Now you can get more out of every conversation with our new follow-up actions, available on every response in your chat! Easily compare different answers from various AI models based on a particular message. Choose follow-up questions to dig deeper into a topic. Or, enhance your generated images with the "Upscale" button for sharper details, or bring them to life with the "Animate" option.

To use follow-up actions, simply click on the three dots '…' next to any message, then select the follow-up options to explore what's possible.

It is easier than ever to share your chat history with others through the new "Share entire chat" option. With just one click, you can generate a unique link to your complete conversation that can be sent to anyone. The share icon located at the top right side of the chat box gives you instant access to this feature.

For those times when you only want to share specific parts of a conversation, the selective message sharing option remains available. This flexibility allows you to choose exactly what content you want others to see, whether it's the full context of your discussion or just the highlights.

We've added an "Undo" function for clearing chat context, giving you a safety net when managing your conversations. Now, after you clear your chat context, an "Undo" button will appear at the top of your chat. This feature allows you to revert the action with a single click, preventing you from losing your previous conversation and avoiding the need to start over.

This helpful feature prevents accidental loss of valuable conversation history, eliminating the frustration of having to start over after an unintended clear action. The simple, one-click recovery process makes managing your chat history more forgiving and user-friendly.

Our new "/repeat" feature lets you receive multiple responses from any bot by typing "/repeat" followed by a number and your prompt. This streamlines your workflow by eliminating the need to manually resubmit prompts, working seamlessly across poe.com and all Poe mobile applications.

The command is particularly powerful when generating images, videos, or audio content, as it lets you explore multiple creative variations simultaneously to find the perfect match for your vision. To try it, simply visit poe.com or open your Poe app and use a command like "/repeat 5 create a watercolor painting of a coastal village at sunset" to see five unique interpretations.

We've recently introduced a convenient feature that allows you to message multiple AI models at once, letting you compare responses from different bots in a single chat thread. This functionality makes it easy to see how various models like GPT-4.5, Claude 3.7 Sonnet, DeepSeek-R1, and others approach the same prompt differently.

To use this feature, simply visit poe.com or any Poe app, @-mention each bot you want to query, add your message, and watch as all the selected models respond. This streamlined approach to model comparison helps you quickly identify which AI best suits your specific needs without switching between multiple conversations.

We've added the ability to send new messages in a chat before a bot has finished responding, eliminating downtime in your AI interactions. This improvement is particularly valuable when working with models that require more processing time, such as reasoning-focused models like DeepSeek-R1 or o3, or video generation tools like Pika and Runway.

As we continue to integrate more powerful AI models throughout the year, this feature will become increasingly essential to your workflow. Now you can maintain productive conversations with multiple questions or requests in sequence, without waiting for each response to complete before moving on to your next thought.