CoolFace
Modelpublic

OrionLLM/GRM-2.6-Opus

sourceHugging Faceapache-2.0updated 2mo agoView on Hugging Face
48likes69downloads
Model Card

<p align="center"> <img src="https://cdn-uploads.huggingface.co/production/uploads/685ea8ff7b4139b6845ce395/_66bkNH630dGeIt2Uuctd.png" alt="logo" width="500"> </p>

<div align="center"> <a href="https://huggingface.co/OrionLLM/GRM-2.6-Opus/" style="text-decoration: none;"> <img src="https://img.shields.io/badge/🤗-HuggingFace-FC926C?style=for-the-badge" alt="HuggingFace"> </a> <a href="https://huggingface.co/collections/OrionLLM/grm-26" style="text-decoration: none;"> <img src="https://img.shields.io/badge/📚-Collection-3B82F6?style=for-the-badge" alt="Collection"> </a> <a href="https://www.apache.org/licenses/LICENSE-2.0" style="text-decoration: none;"> <img src="https://img.shields.io/badge/📜-License-E343BD?style=for-the-badge" alt="License"> </a> </div>

1. Introduction

GRM-2.6-Opus is a merge between OrionLLM/GRM-2.6-Plus and rico03/Qwen3.6-27B-Claude-Opus-Reasoning-Distilled.

GRM-2.6-Opus is a general-purpose AI model optimized for difficult, high-complexity tasks. It is designed to deliver stronger performance for its size while remaining practical, efficient, and accessible for advanced local and research-oriented use.

The model now follows an Opus-style reasoning format, producing more structured, organized, and deliberate reasoning. This merge improves its ability to handle terminal agents, coding workflows, and complex problem-solving tasks, taking advantage of the strong reasoning and agentic capabilities associated with Claude Opus-style distilled behavior.

GRM-2.6-Opus demonstrates improvements over the original GRM-2.6-Plus, especially in structured reasoning, coding, agent workflows, and high-difficulty STEM evaluation.

2. Key Capabilities

  • —Opus-Style Structured Reasoning: GRM-2.6-Opus uses a more organized reasoning format, helping it produce clearer and more reliable solutions for complex tasks.
  • —Improved Terminal Agent Ability: The model is better suited for terminal-based agents, tool-style workflows, debugging, code execution planning, and multi-step technical tasks.
  • —Stronger Coding Performance: The merge improves code reasoning, implementation planning, and difficult programming task handling.
  • —Enhanced General-Purpose Intelligence: GRM-2.6-Opus remains useful across research, STEM, chat, coding, local agents, and advanced problem-solving.
  • —Improved Over GRM-2.6-Plus: The model builds on the original GRM-2.6-Plus and adds stronger structured reasoning behavior through the Opus-style distilled merge.

3. Performance

GRM-2.6-Opus is designed to be a highly capable 27B local AI model for complex reasoning, coding, everyday chat, and agentic workflows. It focuses on delivering better performance for its size, making it a strong option for users who want powerful reasoning without relying only on massive-scale models.

Its core strength is practical intelligence: structured reasoning, strong task understanding, improved coding behavior, stable responses, and the ability to handle difficult problems across multiple domains.

Detailed Benchmarks

<table> <tr> <th style="background: rgba(128,128,128,0.1); text-align: center;">Benchmark</th> <th style="background: rgba(128,128,128,0.1); text-align: center;">GRM-2.6-Opus</th> <th style="background: rgba(128,128,128,0.1); text-align: center;">GRM-2.6-Plus</th> <th style="background: rgba(128,128,128,0.1); text-align: center;">Qwen3.6-27B</th> <th style="background: rgba(128,128,128,0.1); text-align: center;">google/gemma-4-31B-it</th> <th style="background: rgba(128,128,128,0.1); text-align: center;">GPT-5.4-Mini</th> <th style="background: rgba(128,128,128,0.1); text-align: center;">Claude-4.5-Haiku</th> </tr> <tr> <td align="center" colspan="7" style="background: linear-gradient(90deg, rgba(124,58,237,0.45) 0%, rgba(99,102,241,0.42) 50%, rgba(59,130,246,0.45) 100%); font-weight: bold; height:32px; padding-top:2px; padding-bottom:2px;"><i>Knowledge &amp; STEM</i></td> </tr> <tr> <td align="center">GPQA Diamond</td> <td align="center"><b>89.2</b></td> <td align="center">88.3</td> <td align="center">87.8</td> <td align="center">84.3</td> <td align="center">88.0</td> <td align="center">73.0</td> </tr> </table>

<div align="center">

GRM-2.6-Opus is developed by [OrionLLM](https://huggingface.co/OrionLLM) and released under the Apache 2.0 License.

</div>