unconst/Affine-5czsc2fc98-r557-r252-odpo-midrank-longctx-extra-merged

TEXT GENERATIONPricing:Input $0.4 / Cached $0.07 / Output $4Concurrent Unit Cost:3Model Size:35.1BQuant:FP8Context Size:32kTool Calling:SupportedPublished:Aug 16, 2026Architecture:Transformer Featherless Exclusive Cold

The unconst/Affine-5czsc2fc98-r557-r252-odpo-midrank-longctx-extra-merged model is a 35.1 billion parameter language model with a 32768 token context length. It is a LoRA-merged checkpoint derived from `kevin954/Affine-5dfqbbh8ev-sft`. This model is currently a private TTL insurance checkpoint, not yet a formal submission, and is intended for internal evaluation and development.

Loading preview...

Model Overview

This model, unconst/Affine-5czsc2fc98-r557-r252-odpo-midrank-longctx-extra-merged, is a substantial language model featuring 35.1 billion parameters and an extended 32768 token context length. It represents a LoRA-merged checkpoint, specifically derived from the base model kevin954/Affine-5dfqbbh8ev-sft.

Key Characteristics

  • Parameter Count: 35.1 billion, indicating a large-scale model capable of complex language understanding and generation.
  • Context Length: 32768 tokens, allowing for processing and generating very long sequences of text, which is beneficial for tasks requiring extensive context.
  • Origin: It is a LoRA-merged checkpoint, suggesting fine-tuning or adaptation from a pre-existing base model.

Current Status

This particular iteration is described as a "private TTL insurance" checkpoint. This implies it is an internal development or backup version, not yet intended for public release or formal submission. It is likely undergoing further evaluation or refinement before being considered a stable or final version.

When to Consider This Model

Given its current status as a private checkpoint, this model is primarily relevant for:

  • Developers or researchers involved in the unconst or Affine projects.
  • Internal testing and validation within the development team.
  • Exploring the capabilities of large language models with extended context windows in a development environment.