Poe

318 件のポスト

フォロー

Poe

@poe_platform

Fast AI chat, with GPT-4, Claude 2, PaLM 2, DALL-E 3, SDXL and more. At poe.com, or for iOS, Android, MacOS, or Windows at poe.com/download

the internetpoe.com2023年2月からTwitterを利用しています

3 フォロー中

4.9万フォロワー

Poeさんのポスト

固定

Poe

@poe_platform

2月21日

New today: image understanding! We’ve enabled GPT-4v in all GPT-4 conversations, AND added LLaVa-based image understanding for almost all other bots on Poe. You can add images directly to your conversations using the attachment button next to the input box. (1/3)

Image understanding is now available for select bots on Poe!

183

Groq-powered inference for Mixtral is now available on Poe! You can use Mixtral-8x7b-Groq and experience 400 token/second responses. (1/2)

メディアを再生できません。

再読み込み

142

The bot is available at poe.com/Mixtral-8x7b-G and across the Poe iOS, Android, Mac, and Windows apps in the official bots category. (2/2)

Now available: Gemini Pro 1.0 as a base bot! This multimodal bot accepts text, image, and video input, and has been upgraded with overall quality improvements, including improved reasoning, better latency, and multilingual support. We look forward to seeing what you build!

メディアを再生できません。

再読み込み

132

Just launched: Playground v2.5 on Poe! This new image model produces more realistic and visually compelling images, with key improvements in color vibrancy and contrast, finer human-related details, and multi-aspect ratio support. (1/3)

Playground v2.5 is now available on Poe!

158

Playground’s internal evaluation and user studies have found that v2.5 demonstrates a significant increase in aesthetic quality and outperforms almost all other models. (2/3)

Playground v2.5 is available now at poe.com/Playground-v2.5 and across all Poe apps. (3/3)

Mistral Large, released by Mistral this morning, is now available to all users on Poe! This is the strongest Mistral model available so far and Poe is the easiest place to use it across all platforms: web, iOS, Android, MacOS, and Windows. (1/2)

Mistral Large, the largest and most state-of-the-art model from Mistral AI, is now available on Poe. Poe is the easiest way to interact with Mistral Large.

200

We are still in the process of evaluating it, but it appears to approach GPT-4, Claude 2, and Gemini in terms of quality, while including its own strengths, weaknesses, and biases. You can try it at: poe.com/Mistral-Large. Enjoy!

Now on Poe: Gemma 7b-FW! Google's new lightweight model is built from the same research and technology as the Gemini models, and outperforms similar sized open models in general language understanding, reasoning, math, and coding, according to their internal evaluations. (1/2)

133

Gemma-7b-FW is hosted by

@FireworksAI_HQ

. It’s available now at poe.com/Gemma-7b-FW and across all Poe apps. (2/2)

Reka Flash from

is now available on Poe! Reka Flash is a state-of-the-art 21B multimodal language model that works with text, image, and short video input. Reka Flash is highly capable and very performant for its size. (1/2)

You can try Reka Flash today at poe.com/RekaFlash or via the Poe apps available on all platforms.

183

Image understanding works well for a variety of tasks. For instance, you can upload a photo and ask Assistant to describe it, attach a graph and have GPT-4 explain the data, or even add a picture of a dish and ask Mistral Medium to share the recipe for it. (2/3)

We're excited to see how you use image understanding. It's available immediately on web and across all Poe apps on all platforms! (3/3)

Code-Llama-70B-FW is now available on Poe! Hosted by

@FireworksAI_HQ

, CodeLlama-70B is the largest and best-performing model in the Code Llama family, and one of the highest performing open source code models available. (1/2)

123

Code-Llama-70B-FW is available at poe.com/Code-Llama-70B and across all Poe apps. Enjoy! (2/2)

Code-Llama-70B-FW - Poe

Image Remix is now available on iOS! Create custom versions of any generated or uploaded image simply by modifying the prompt. Give it a try on any bot based on SDXL or Playground-v2 and share what you create!

メディアを再生できません。

再読み込み

We’ve added Mixtral 8x7B as a base model for bot creation! All Poe users can now create prompt bots on Mixtral-8x7B-Chat, a chat fine-tuned version of one of the strongest open source models. We can’t wait to see what you build!

メディアを再生できません。

再読み込み

126

New on Poe: Image remix! Customize the style, layout, colors and other aspects of a generated or attached image by editing the prompt given to any bot based on SDXL or Playground v2! (1/3)

メディアを再生できません。

再読み込み

For example: let’s say you generate an image of two cute kittens, but now you want to change the background. By clicking ‘Remix’ and altering the prompt, you’re able to make the desired changes to your image while maintaining the consistency from the original generation. (2/3)

Image remix is currently available in all image generation bots based on SDXL and Playground v2 on Web, macOS, and Windows, with support for iOS and Android coming soon. Enjoy! (3/3)

3,889

Xを使ってみよう

今すぐ登録して、タイムラインをカスタマイズしましょう。

Appleのアカウントで登録

アカウントを作成

アカウントを登録することにより、利用規約とプライバシーポリシー（Cookieの使用を含む）に同意したとみなされます。

トレンドはありません。