01-ai/Yi-1.5-34B-Chat-16K

TEXT GENERATIONConcurrency Cost:2Model Size:34BQuant:FP8Ctx Length:32kPublished:May 15, 2024License:apache-2.0Architecture:Transformer0.0K Open Weights Cold

The Yi-1.5-34B-Chat-16K model by 01-ai is an upgraded 34 billion parameter large language model with a 16K context window. Continuously pre-trained on 500 billion tokens and fine-tuned on 3 million diverse samples, it delivers enhanced performance in coding, mathematics, and reasoning. This model maintains strong capabilities in language understanding and instruction-following, making it suitable for complex conversational AI applications.

Loading preview...

Yi-1.5-34B-Chat-16K Overview

Yi-1.5 is an upgraded series of large language models developed by 01-ai, building upon the original Yi models. The Yi-1.5-34B-Chat-16K is a 34 billion parameter chat-optimized variant featuring an extended context length of 16,384 tokens. This model underwent continuous pre-training on an additional 500 billion high-quality tokens and was fine-tuned using 3 million diverse samples.

Key Capabilities

  • Enhanced Performance: Compared to its predecessor, Yi-1.5 demonstrates stronger capabilities in coding, mathematical problem-solving, and general reasoning tasks.
  • Instruction Following: It exhibits improved instruction-following abilities, making it more effective for conversational and task-oriented applications.
  • Core Language Skills: The model retains excellent performance in fundamental areas such as language understanding, commonsense reasoning, and reading comprehension.
  • Extended Context: The 16K context window allows for processing and generating longer, more complex interactions and documents.

Good For

  • Complex Conversational AI: Its improved reasoning and instruction-following, combined with a large context window, make it suitable for advanced chatbots and virtual assistants.
  • Code Generation & Analysis: The enhanced coding performance suggests utility in developer tools and programming assistance.
  • Mathematical & Logical Tasks: Stronger mathematical and reasoning capabilities are beneficial for applications requiring precise logical deductions.