KIEFERSA/Sophea-Genesis-1

VISIONConcurrent Unit Cost:2Model Size:27BQuant:FP8Context Size:32kTool Calling:SupportedPublished:Jul 12, 2026License:apache-2.0Architecture:Transformer0.0K Open Weights Featherless Exclusive Cold

KIEFERSA/Sophea-Genesis-1 is a 27 billion parameter language model developed by Kiefer SA, based on Qwen3.6-27B. It is specifically optimized for generating well-formed Modern Greek, achieving leading Greek morphological accuracy among comparable open models. While excelling in Greek inflection, it retains broad competence in both Greek and English for general knowledge, reading comprehension, and instruction following. This model is ideal for applications requiring high-quality Greek output, particularly in environments with data privacy constraints.

Loading preview...

Sophea-Genesis-1: Greek Morphology and General Competence

Sophea-Genesis-1 is a 27-billion-parameter open-weight language model from Kiefer SA, built upon the Qwen3.6-27B architecture. Its primary distinction lies in its leading accuracy for Modern Greek inflectional morphology among comparably sized open models. This capability is crucial for generating grammatically correct Greek, a highly inflected language.

Key Capabilities

  • Superior Greek Morphology: Achieves the highest accuracy in both recognition and generation of correctly inflected Greek word-forms on the MORFES benchmark, outperforming other general-purpose and Greek-specific open models.
  • Retained General Competence: Maintains strong performance in general-capability benchmarks for knowledge, reading comprehension, and instruction following in both Greek and English, comparable to or exceeding similarly sized models.
  • Open Weights & On-Premise Deployment: Released under the Apache-2.0 license, allowing for commercial use, modification, and deployment on private infrastructure, ensuring data privacy and eliminating per-request fees.

Good for

  • Greek-language applications where high-quality, grammatically precise Greek output is essential.
  • Organizations with strict data privacy requirements (e.g., government, law, finance, healthcare) that need to process Greek text in-house.
  • Educational and language-learning tools focused on presenting accurate Greek word-forms and inflectional paradigms.
  • As a base model for further fine-tuning on specific Greek domains, with the MORFES benchmark available to verify morphological accuracy preservation.