Saxo/Linkbricks-Horizon-AI-Japanese-Superb-V2-70B

TEXT GENERATIONPricing:Input $3.5 / Cached $0.7 / Output $8.3Concurrent Unit Cost:4Model Size:70BQuant:FP8Context Size:8kTool Calling:SupportedPublished:Dec 18, 2024License:apache-2.0Architecture:Transformer Open Weights Featherless Exclusive Cold

Saxo/Linkbricks-Horizon-AI-Japanese-Superb-V2-70B is a 70 billion parameter Japanese-enhanced language model developed by Yunsung Ji (Saxo) at Linkbricks. It is fine-tuned from Saxo/Linkbricks-Horizon-AI-Llama-3.3-Japanese-70B-Advanced using SFT and DPO on 30 million Japanese news and wiki corpora. This model excels in high-dimensional analysis of customer reviews and social posts, coding, writing, mathematics, and complex logical reasoning, with cross-language capabilities in Japanese, Korean, Chinese, and English, and supports a 128k context window and function calling.

Loading preview...

Overview

Saxo/Linkbricks-Horizon-AI-Japanese-Superb-V2-70B is a 70 billion parameter language model developed by Yunsung Ji (Saxo), a data scientist at Linkbricks, an AI and big data analytics company. This model is an advanced Japanese-enhanced variant, fine-tuned using Supervised Fine-Tuning (SFT) and Direct Preference Optimization (DPO) from the Saxo/Linkbricks-Horizon-AI-Llama-3.3-Japanese-70B-Advanced base model. It leverages 30 million Japanese news and wiki corpora, alongside cross-language training data for Japanese, Korean, Chinese, and English, and specialized data for mathematics and logical reasoning.

Key Capabilities

  • Enhanced Japanese Processing: Optimized for deep understanding and generation in Japanese.
  • Multilingual Cross-Learning: Strengthened with cross-training data across Japanese, Korean, Chinese, and English.
  • Advanced Analytical Skills: Excels in high-dimensional analysis of customer reviews and social media posts.
  • Coding and Writing: Improved performance in code generation and various writing tasks.
  • Mathematical and Logical Reasoning: Capable of handling complex mathematical problems and intricate logical judgments.
  • Extended Context Window: Supports a 128k context window for processing longer inputs.
  • Function Calling: Includes support for function calling and tool calling.

Good for

  • Applications requiring sophisticated Japanese language understanding and generation.
  • Cross-lingual tasks involving Japanese, Korean, Chinese, and English.
  • Analyzing large volumes of customer feedback and social media data.
  • Development of AI assistants for coding, content creation, and technical writing.
  • Solving complex problems in mathematics and logic.

Finetunes
1 models