FRPO/qwen3-1.7b-a12_dapo_frpo-k1-cNone-clip0.2_0.28-mb4-eta100-bs256x5-n2

TEXT GENERATIONPricing:Input $0.32 / Cached $0.064 / Output $1.6Concurrent Unit Cost:1Model Size:2BQuant:BF16Context Size:32kTool Calling:SupportedPublished:Aug 15, 2026Architecture:Transformer Featherless Exclusive Cold

FRPO/qwen3-1.7b-a12_dapo_frpo-k1-cNone-clip0.2_0.28-mb4-eta100-bs256x5-n2 is a 2 billion parameter language model based on the Qwen3-1.7B architecture. This model has been fine-tuned using Reinforcement Learning (RL) with the FRPO method as part of the KL-in-LLM-RL experiments. It is specifically designed for research and development in RL fine-tuning techniques for large language models.

Loading preview...

Overview

This model, FRPO/qwen3-1.7b-a12_dapo_frpo-k1-cNone-clip0.2_0.28-mb4-eta100-bs256x5-n2, is a 2 billion parameter checkpoint derived from the Qwen/Qwen3-1.7B base model. It was developed as part of the KL-in-LLM-RL / FRPO experimental series, utilizing the verl framework for its Reinforcement Learning (RL) fine-tuning process.

Key Characteristics

  • Base Model: Qwen3-1.7B, a causal language model.
  • Fine-tuning Method: Reinforcement Learning (RL) using the FRPO (Fast Reinforcement Learning with Policy Optimization) approach.
  • Training Framework: Implemented with the verl library.
  • Checkpoint: The repository contains the global_step_200 checkpoint.
  • Weights: Provided in fp32 safetensors format, directly as saved by the trainer without further post-processing.
  • Configuration: The specific run configuration parameters are encoded within the model's repository name.

Intended Use Cases

This model is primarily suited for:

  • RL Research: Experimentation and development in Reinforcement Learning fine-tuning techniques for large language models.
  • Comparative Studies: Analyzing the impact of FRPO and similar RL methods on model performance and behavior.
  • Academic Research: Investigating the properties and capabilities of RL-tuned Qwen3-1.7B variants.

Finetunes
1 models