Select your cookie preferences

We use essential cookies and similar tools that are necessary to provide our site and services. We use performance cookies to collect anonymous statistics, so we can understand how customers use our site and make improvements. Essential cookies cannot be deactivated, but you can choose “Customize” or “Decline” to decline performance cookies.

If you agree, AWS and approved third parties will also use cookies to provide useful site features, remember your preferences, and display relevant content, including relevant advertising. To accept or decline all non-essential cookies, choose “Accept” or “Decline.” To make more detailed choices, choose “Customize.”
Sign in
Sign in
or
Create a new account
About
Categories
Delivery Methods
Solutions
Resources
 Your Saved List Become a Channel Partner Sell in AWS Marketplace Amazon Web Services Home Help

Amazon Sagemaker

Amazon SageMaker is a fully-managed platform that enables developers and data scientists to quickly and easily build, train, and deploy machine learning models at any scale. With Amazon SageMaker, all the barriers and complexity that typically slow down developers who want to use machine learning are removed. The service includes models that can be used together or independently to build, train, and deploy your machine learning models.

product logo

Solar Mini Chat - Quant Free trial

By:
Upstage
Compact, yet powerful LLM by Upstage
  • This product has been removed and is no longer available to new customers.

Product Overview

This is quantized version of Solar Mini Chat. Solar Mini Chat is a compact yet powerful advanced English/Korean large language model. It is specifically fine-tuned for multi-turn chat purposes, demonstrating enhanced performance across a wide range of natural language processing tasks. This fine-tuning equips it with the ability to handle extended conversations more effectively, making it particularly adept for interactive applications. It employs a scaling method called ‘depth up-scaling’ (DUS), which is comprised of depthwise scaling and continued pretraining. DUS allows for a much more straightforward and efficient enlargement of smaller models than other scaling methods such as mixture-of-experts.

Key Data

By
Upstage
Categories
Natural Language Processing
Generative AI
Generation-Text
Type
Model Package
Fulfillment Methods
Amazon SageMaker

Highlights

  • Key Features

    • High Performance: Solar Mini Chat is specially optimized for chat applications in Korean. It surpasses many open-source models, especially in handling the intricacies of Korean language interactions, offering more natural and accurate responses.
    • Enhanced User Experience: Solar Mini Chat advanced grasp of Korean language nuances significantly elevates user interactions in chat environments. It provides precise responses to user inputs, ensuring clearer communication and more efficient problem resolution in Korean chat applications.

  • Key Applications

    The capabilities of Solar Mini Chat are particularly valuable in scenarios requiring a nuanced understanding and responsive execution of complex conversational exchanges, particularly in the Korean language. With its advanced ability to interpret and engage in detailed dialogues, Solar Mini Chat is an exceptional choice for applications demanding interactive, context-aware communication, especially tailored to the subtleties and intricacies of the Korean language.

  • Key Tasks

    • Instruction Following
    • Natural Language Understanding
    • Natural Language Generation
    • Question Answering
    • Text Generation and Editing
    • Translation Services
    • Summarization
    • Information Extraction
    • Chatbots and Conversational Interfaces
    • Educational and Learning Assistance
    • Data Analysis and Report Writing
    • Code Generation and Debugging
    • Content Recommendation Systems
    • Sentiment Analysis and Customer Feedback Processing

Not quite sure what you’re looking for? AWS Marketplace can help you find the right solution for your use case. Contact us

Usage Information

Model input and output details

Input

Summary

We support request payload compatible to OpenAI's Chat completion endpoint.

Limitations for input type
Solar support maximum 4k(4096) context for input and generated tokens.
Input MIME type
application/json
Sample input data
{
    "messages": [
      {
        "role": "system",
        "content": "You are a helpful assistant."
      },
      {
        "role": "user",
        "content": "Can you provide a Python script to merge two sorted lists?"
      }
    ],
    "temperature": 0.7,
}
Expand all input descriptions
Field name
messages
Description
List of messages that contains role and content. Role must be one of [system, user, assistant].
Constraints

Type: FreeText Limitations: You must provide list of messages that contains role and content. Role must be one of [system, user, assistant].

Required
Yes
Field name
model
Description
You can use model parameter for compitability, but since We have only one model, this parameter is not required.
Constraints

Default value: /opt/ml/model Type: FreeText

Required
No
Field name
frequency_penalty
Description
Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim.
Constraints

Default value: 0.0 Type: Continuous Minimum: -2.0 Maximum: 2.0

Required
No
Field name
presence_penalty
Description
Number between -2.0 and 2.0. Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics.
Constraints

Default value: 0.0 Type: Continuous Minimum: -2.0 Maximum: 2.0

Required
No
Field name
max_tokens
Description
The maximum number of tokens that can be generated in the chat completion. Solar support maximum 4k(4096) context for input and generated tokens.
Constraints

Default value: 16 Type: Integer Minimum: 0 Maximum: 4096

Required
No
Field name
temperature
Description
What sampling temperature to use, between 0 and 2. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic.
Constraints

Default value: 1.0 Type: Continuous Minimum: 0.0 Maximum: 2.0

Required
No
Field name
top_p
Description
An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.
Constraints

Default value: 1.0 Type: Continuous Minimum: 0.0 Maximum: 1.0

Required
No
Field name
stream
Description
Specifies whether to stream the response.
Constraints

Default value: false Type: Categorical Allowed values: true, false

Required
No

Output

Summary

We support response payload compatible to OpenAI's Chat completion endpoint.

Output MIME type
application/json
Sample output data
{
  "id": "cmpl-cc3f71b58086441fa6a53480b94fd0ec",
  "object": "chat.completion",
  "created": 3421,
  "model": "/opt/ml/model",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "What do you call a fake noodle?\n\nAn impasta!"
      },
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 34,
    "total_tokens": 334,
    "completion_tokens": 300
  }
}
Expand all output descriptions
Field name
id
Description
An ID of the completion. Type: FreeText
Always returned
Yes
Data type
Field name
object
Description
The type of completion object. Always chat.completion. Type: FreeText
Always returned
Yes
Data type
Field name
created
Description
The number of tokens generated from Solar. Type: Integer
Always returned
Yes
Data type
Field name
model
Description
The name of the model used by completion. Type: FreeText
Always returned
Yes
Data type
Field name
choices
Description
A list of chat completions, including index, message, finish_reason. Type: FreeText
Always returned
Yes
Data type
Field name
usage
Description
The number of tokens used for generating completion, including prompt_tokens, total_tokens, and completion_tokens. Type: Integer
Always returned
Yes
Data type

Sample notebook

Additional Resources

End User License Agreement

By subscribing to this product you agree to terms and conditions outlined in the product End user License Agreement (EULA)

Support Information

Solar Mini Chat - Quant

Contact us for model fine-tuning request. https://www.upstage.ai/llm/custom-model-request

AWS Infrastructure

AWS Support is a one-on-one, fast-response support channel that is staffed 24x7x365 with experienced and technical support engineers. The service helps customers of all sizes and technical abilities to successfully utilize the products and features provided by Amazon Web Services.

Learn More

Refund Policy

We do not support any refunds currently.

Customer Reviews

There are currently no reviews for this product.
View all