d4rkninja/tanpo-retention
The d4rkninja/tanpo-retention model is a compact 1.2 billion parameter language model, fine-tuned from LiquidAI/LFM2.5-1.2B-Instruct, with a 32,768-token context length. Developed by d4rkninja, this specialized model is designed for customer retention workflows, including onboarding, churn-save, renewals, and customer success playbooks. It excels at generating structured responses for specific business scenarios, making it suitable for local and inexpensive deployment in domain-specific applications.
Loading preview...
Tanpo Retention: A Specialized Customer Retention LLM
Tanpo Retention is a compact, 1.2 billion parameter language model developed by d4rkninja, specifically fine-tuned for customer retention and success workflows. Built upon the LiquidAI/LFM2.5-1.2B-Instruct base model, it offers a substantial 32,768-token context length, making it suitable for detailed business scenarios. This model is part of the Tanpo family of domain-specialized models, optimized for local and inexpensive deployment rather than general-purpose tasks.
Key Capabilities
- Customer Retention Workflows: Specializes in generating structured responses for onboarding, activation, churn-save, winback, and renewal processes.
- Customer Success Playbooks: Assists with health-score follow-up, expansion strategies, and escalation drafts.
- Compact and Efficient: Designed for edge or on-device deployment, enabling private and low-cost inference.
- Performance: Achieved a 96.0% overall score on internal domain evaluation, showing a +1.8 percentage point improvement over its base model, with significant gains in escalation, renewals, and winback.
Best For
- Drafting customer-success playbooks and renewal preparation plans.
- Creating onboarding and activation plans tied to time-to-value.
- Generating churn-save, winback, expansion, and escalation drafts.
- Interpreting health scores and developing practical account action plans.