unconst/Affine-5czsc2fc98-r214-merged

TEXT GENERATIONPricing:Input $0.4 / Cached $0.07 / Output $4Concurrent Unit Cost:3Model Size:35.1BQuant:FP8Context Size:32kTool Calling:SupportedPublished:Aug 14, 2026Architecture:Transformer Featherless Exclusive Cold

Affine-5czsc2fc98-r214-merged is a 35.1 billion parameter model developed by unconst, derived from a LoRA merge of kevin954/Affine-5dfqbbh8ev-sft. This model is specifically designed for private TTL insurance applications, indicating a specialized focus on domain-specific tasks within the insurance sector. Its primary differentiator lies in its targeted application for internal insurance processes rather than general-purpose language generation.

Loading preview...

Model Overview

The unconst/Affine-5czsc2fc98-r214-merged model is a 35.1 billion parameter language model. It was created by unconst through a LoRA merge from the kevin954/Affine-5dfqbbh8ev-sft base model. This particular iteration, identified as r214-merged, is an internal checkpoint.

Key Characteristics

  • Parameter Count: 35.1 billion parameters.
  • Context Length: Supports a context window of 32,768 tokens.
  • Origin: LoRA-merged from kevin954/Affine-5dfqbbh8ev-sft.
  • Development Stage: Described as a "merged checkpoint salvage" and "not a submission until Stage-5 gate clears," indicating it is an ongoing development or internal version.

Intended Use Case

This model is explicitly noted for "Private TTL insurance." This suggests a highly specialized application within the insurance industry, likely for tasks such as:

  • Processing and analyzing insurance-related documents.
  • Assisting with internal insurance workflows.
  • Handling domain-specific queries related to TTL (Term Life Insurance) policies.

Due to its specialized nature and internal development status, it is not intended for general-purpose use or public deployment at this stage.

Finetunes
1 models