Local-Novel-LLM-project/Ninja-v1

TEXT GENERATIONConcurrency Cost:1Model Size:7BQuant:FP8Ctx Length:4kPublished:Apr 24, 2024License:apache-2.0Architecture:Transformer0.0K Open Weights Cold

Ninja-v1 is a 7 billion parameter large language model developed by Local-Novel-LLM-project, fine-tuned from Mistral-7B-v0.1. It is specifically designed for high-quality generation in both Japanese and English, featuring enhanced memory capabilities for long-context conversations. This model excels at maintaining coherence over extended interactions, making it suitable for applications requiring sustained dialogue or narrative generation.

Loading preview...

Ninja-v1: A Mistral-7B Fine-tune for Bilingual Generation

Ninja-v1 is a 7 billion parameter large language model developed by Local-Novel-LLM-project, built upon the Mistral-7B-v0.1 architecture. This model distinguishes itself through its dual focus on high-quality Japanese and English text generation, a capability achieved through fine-tuning on a novel dataset. A key feature of Ninja-v1 is its improved "memory ability," designed to maintain context and coherence even during long-context generation, addressing a common challenge in LLMs.

Key Capabilities & Features

  • Bilingual Proficiency: Optimized for generating high-quality text in both Japanese and English.
  • Enhanced Context Retention: Designed to remember information and maintain coherence over extended conversations or long-form text generation.
  • Mistral-7B Base: Leverages the robust architecture of Mistral-7B-v0.1.
  • Multi-turn Conversation Support: Adopts the Vicuna prompt format, facilitating natural multi-turn interactions.
  • Merge-based Development: Created by merging models like WizardLM2 (multitasking), Antler-7B (novel writing), and NTQAI/chatntq-ja-7b-v1.0 (Japanese specialization).

Ideal Use Cases

  • Long-form Content Generation: Suitable for tasks requiring sustained narrative or dialogue, such as creative writing or interactive storytelling.
  • Bilingual Applications: Excellent for scenarios needing fluent text generation in both Japanese and English.
  • Conversational AI: Its multi-turn conversation support and memory capabilities make it well-suited for chatbots and virtual assistants that require extended interactions.