RaspizdAI/tolk-7B

TEXT GENERATIONPricing:Input $0.4 / Cached $0.08 / Output $0.8Concurrent Unit Cost:1Model Size:7.6BQuant:FP8Context Size:32kTool Calling:SupportedPublished:Sep 19, 2026License:mitArchitecture:Transformer Open Weights Featherless Exclusive Cold

tolk-7B is a 7.6 billion parameter language model developed by RaspizdAI, based on the DeepSeek-R1-Distill-Qwen-7B architecture. This model is specifically fine-tuned for multi-turn conversational dialogue, persona alignment, and structured text generation. It supports both Russian and English languages, making it suitable for bilingual conversational AI applications.

Loading preview...

Model Overview

tolk-7B is a 7.6 billion parameter language model developed by RaspizdAI, built upon the DeepSeek-R1-Distill-Qwen-7B base architecture. It is designed for robust performance in conversational AI tasks, supporting both Russian and English languages.

Key Capabilities

  • Multi-turn Conversational Dialogue: Excels at maintaining coherent and contextually relevant conversations over multiple exchanges.
  • Persona Alignment: Capable of generating responses that adhere to a specified persona or character.
  • Structured Text Generation: Can produce text outputs following defined formats or structures.
  • Bilingual Support: Optimized for tasks in both Russian (ru) and English (en).

Good For

  • Developing chatbots and virtual assistants requiring nuanced conversational abilities.
  • Applications needing persona-driven dialogue generation.
  • Generating structured text in a conversational context.
  • Bilingual conversational systems in Russian and English.