Arcee AI Unveils Arcee-Spark: A Game-Changing Qwen2-Based Custom Model

PRESS RELEASE
Published June 28, 2024

Arcee AI has launched Arcee-Spark, their first custom model based on Qwen2 7B, demonstrating impressive potential. This new model outperforms Meta’s Llama 3 8B on AGIEval and OpenAI’s GPT-3.5 on MT-Bench, setting a new standard for small language models.

Key Features of Arcee-Spark:

  • Fine-Tuning: Qwen2 Base fine-tuned on 1.8 million samples
  • Model Merging: Combined with Qwen2-7B-Instruct using MergeKit
  • Post-Training: Further enhanced through Direct Policy Optimization (DPO)
  • Performance Metrics:
  • AGIEval: 51.11
  • MT-Bench: 8.46
  • BigBenchHard: 45.78
  • EQ-Bench: 71.4
  • Licensing: Released under Apache 2.0 license

The Evolution of Small LLMs

It’s astounding how far small language models have progressed. The Qwen2 7B is now able to match GPT-3.5, a model that amazed us less than two years ago with ChatGPT.

The post Arcee AI Unveils Arcee-Spark: A Game-Changing Qwen2-Based Custom Model first appeared on .

Binary News Network

Binary News Network is a Content Syndication Platform that allows businesses or proprietary newswires to bring visibility to their content by syndicating it to premium, high-visibility networks and sites and drive visibility to your online content.