INSAIT-Institute/BatchOnline-GRPO-7B

TEXT GENERATIONPricing:Input $0.4 / Cached $0.08 / Output $0.8Concurrent Unit Cost:1Model Size:7.6BQuant:FP8Context Size:32kTool Calling:SupportedPublished:Aug 12, 2026License:cc-by-nc-sa-4.0Architecture:Transformer Open Weights Featherless Exclusive Cold

INSAIT-Institute/BatchOnline-GRPO-7B is a 7.6 billion parameter code verifier model developed by INSAIT-Institute, fine-tuned from DeepSeek-R1-Distill-Qwen-7B. This model is trained with a semi-online Generative Reinforcement Policy Optimization (GRPO) approach, where the generator policy is synced every four steps. It excels at judging and ranking candidate solutions for competitive programming problems, making it suitable for automated code evaluation and as a reward function in RLHF/RLAIF for code generation.

Loading preview...

Aletheia: BatchOnline-GRPO-7B Code Verifier

BatchOnline-GRPO-7B is a 7.6 billion parameter code verifier developed by INSAIT-Institute, part of the Aletheia project. It is fine-tuned from deepseek-ai/DeepSeek-R1-Distill-Qwen-7B using the INSAIT-Institute/Aletheia-Train dataset. This model employs a semi-online GRPO training method, where the generator policy is synchronized every four steps, balancing performance and computational cost in Reinforcement Learning with Verifiable Rewards (RLVR) pipelines.

Key Capabilities

  • Code Verification: Judges and ranks candidate solutions for competitive programming problems, identifying correct code snippets from a set of options.
  • Semi-Online Training: Utilizes a specific GRPO variant that updates the generator policy periodically, offering a practical approach to RLVR for code.
  • Multi-language Support: Trained on problems and solutions in Python, C++, and Java, enabling broad applicability across programming languages.
  • Research Testbed: Part of a larger effort to analyze the impact of thinking traces, negative samples, and on-policy training in code verifier development.

Good For

  • RLHF / RLAIF: Serving as a plug-and-play reward function to optimize code generation policies.
  • Automated Evaluation: Acting as an LLM-as-a-judge for various code-related tasks, providing objective assessments.
  • Research: Investigating the efficacy of different training recipes for robust and efficient code verifiers, particularly concerning the trade-offs between performance and cost.

Finetunes
1 models