FRPO/qwen3-1.7b-a9_dapo-dapo-noKL-clip0.2_0.28-mb4-eta100-bs256x5-n2-r3

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-a9_dapo-dapo-noKL-clip0.2_0.28-mb4-eta100-bs256x5-n2-r3 is a 2 billion parameter language model based on the Qwen3-1.7B architecture, featuring a 32768-token context length. This model is an RL fine-tuned checkpoint from the KL-in-LLM-RL / FRPO experiments, specifically optimized using the verl framework. It is designed for research into reinforcement learning applications for large language models.

Loading preview...

Model Overview

FRPO/qwen3-1.7b-a9_dapo-dapo-noKL-clip0.2_0.28-mb4-eta100-bs256x5-n2-r3 is a 2 billion parameter model derived from the Qwen/Qwen3-1.7B base architecture, featuring a substantial 32768-token context window. This particular checkpoint is a result of Reinforcement Learning (RL) fine-tuning conducted within the KL-in-LLM-RL / FRPO experimental framework.

Key Characteristics

  • RL Fine-tuned: This model has undergone specific reinforcement learning optimization using the verl framework, indicating a focus on improving performance through iterative learning from feedback.
  • Base Model: Built upon the robust Qwen3-1.7B foundation, suggesting strong general language understanding capabilities prior to RL tuning.
  • Parameter Count: With 2 billion parameters, it offers a balance between computational efficiency and performance for various tasks.
  • Context Length: A 32768-token context window allows for processing and generating longer sequences of text, beneficial for complex tasks requiring extensive context.

Intended Use Cases

This model is primarily suited for:

  • RL Research: Ideal for researchers exploring the impact and effectiveness of reinforcement learning techniques on large language models.
  • Experimental Applications: Users interested in evaluating models fine-tuned with specific RL configurations, as indicated by the detailed run configuration encoded in its name.
  • Prototyping: Suitable for developing and testing applications where a Qwen3-based model with RL enhancements might offer unique advantages.

Finetunes
1 models