Used to check for browser translation.
用于检测浏览器翻译。
ブラウザの翻訳を検出する

AIGNE 0.4: Supports GPT-4o mini model & AI service pricing update

Yechao
Jul 19, 2024 · edited
B
Blogs
cover

With more and more developers joining our AIGNE platform👏, we are continuously committed to improving their user experience and providing more affordable AI services. Therefore, we have updated the pricing of AI services💰. At the same time, we have quickly supported the latest OpenAI GPT-4o mini model, which is not only cheaper but also faster in response🚀. We have also introduced some new features and bug fixes, and we welcome everyone to experience them!

Pricing update#

Following the latest round of pricing adjustments, our AI service model rates have significantly decreased, with reductions of up to 90% 🎉🎉🎉! This adjustment aims to allow developers to use our advanced models at a lower cost and gain more value. Now, you can experience our high-quality AI services at more competitive prices!

Model (ABT/1K Tokens)

Input (Original)

Output (Original)

Input 🆕

Output 🆕

gpt-3.5-turbo

0.0025

0.005

0.00025

0.00075

gpt-3.5-turbo-16k

0.0025

0.005

0.00025

0.00075

gpt-3.5-turbo-instruct

0.00375

0.005

0.00075

0.001

gpt-4

0.075

0.15

0.015

0.03

gpt-4-32k

0.15

0.3

0.03

0.06

gemini-pro

0.0025

0.005

0.0025

0.0075

mistralai/mistral-7b-instruct

-

-

0.00003

0.00003

mistralai/mixtral-8x7b-instruct

-

-

0.00012

0.00012

dall-e-2

-

0.05

-

0.01

dall-e-3

-

0.3

-

0.06

text-embedding-ada-002

0.05

-

0.00005

-

gpt-4o

0.075

0.15

0.0025

0.0075

gpt-4o-mini 🆕

-

-

0.000075

0.0003

New Features#

GPT-4o mini model#

OpenAI recently released the latest GPT-4o mini model, and AINGE now supports this model🚀

OpenAI is committed to making intelligence as broadly accessible as possible. Today, we're announcing GPT-4o mini, our most cost-efficient small model. We expect GPT-4o mini will significantly expand the range of applications built with AI by making intelligence much more affordable. GPT-4o mini scores 82% on MMLU and currently outperforms GPT-4 on chat preferences in LMSYS leaderboard. It is priced at 15 cents per million input tokens and 60 cents per million output tokens, an order of magnitude more affordable than previous frontier models and more than 60% cheaper than GPT-3.5 Turbo.

You can choose the new GPT4o mini model in the model settings of any large language model Agent in AIGNE Studio (of course, you can also set the default model to GPT4o mini in the model settings of the project, so that newly created Agents will default to using this model).

Screenshot 2024-07-19 at 08.24.21.png

Agent Combination#

AIGNE now supports the combination and encapsulation of Agents. This means that an Agent can call one or more other Agents, achieving the following two powerful functionalities:

  1. Predefined parameters: You can pre-set the parameters of the base Agent so that they are automatically applied when called.
  2. Parallel Invocation and Output Combination: You can call multiple agents simultaneously and combine their outputs to obtain a more comprehensive result.

Screenshot 2024-07-19 at 09.22.36.png
1. Switch processing to Call Agent2. Add Agent to call3. Pick up outputs from called Agents

Prompt editor variable experience enhancement#

Optimize the user experience of the Prompt editor, now you can easily insert a variable by entering {, while also optimizing automatic synchronization when renaming input.

Kapture 2024-07-19 at 11.24.28.gif


Support sharing Agent Output to the ArcBlock community.#

AINGE now supports sharing an agent output to ArcBlock community.

  1. Set the share output selection for agent in AIGNE Studio to the ArcBlock Community option

image.png
Choose ArcBlock Community optionin the share output settings

  1. The ArcBlock Community button will be displayed in the Agent application output, click to share to ArcBlock Community

Screenshot 2024-07-19 at 10.25.28.png
Share to ArcBlock Community

Other features and bug fixes#

  • Chat input box optimization: Support Shift+Enter to input multiple lines of text during chat
  • OpenGraph optimization: Now supports using the avatar and name in the Agent appearance settings as OpenGraph data display
  • Language selection optimization: support language search, while displaying flags in the options to make it easier for users to choose
  • Agent icon optimization: support using the avatar configured by the Agent as the application icon

Update Notes: Thank you for your feedback#

We believe that the core of our product lies in a vibrant user community. Your insights and suggestions are extremely valuable to us, and we are committed to incorporating your feedback to make AIGNE even better. In this release, we focus on the latest updates and bug fixes inspired by you (our users). From minor adjustments to major changes, every update is a step towards enhancing your experience and meeting your needs. Thank you for your continued support and involvement. Let's see how your feedback has helped us improve the content in this version!

Help us improve 🌟#

We hope you have a better user experience in the new version of AIGNE! 🚀

Heartfelt thanks to every user for your attention and support. We will continue to improve to better meet your needs. If you have any requirements, suggestions, or problems while using it, please provide feedback in our community, and we will do our best to help. 💬


2.0.154