ahammad115566/qwen-smeft

TEXT GENERATIONPricing:Input $0.468 / Output $1.82Concurrent Unit Cost:1Model Size:8BQuant:FP8Context Size:32kTool Calling:SupportedPublished:Jul 31, 2026License:apache-2.0Architecture:Transformer0.0K Open Weights Featherless Exclusive Cold

ahammad115566/qwen-smeft is a domain-adapted large language model fine-tuned from Qwen3-8B by Ahmed Hammad and Veronica Sanz. Optimized for Standard Model Effective Field Theory (SMEFT), it excels at SMEFT operator reasoning, EFT basis translation, and physics-aware scientific dialogue. This model is specifically designed for structured theoretical question answering within the particle physics domain.

Loading preview...

Overview

This model, ahammad115566/qwen-smeft, is a domain-adapted large language model (LLM) specifically fine-tuned for Standard Model Effective Field Theory (SMEFT). Developed by Ahmed Hammad and Veronica Sanz, it is based on the Qwen3-8B architecture and was trained using LoRA on a curated corpus of SMEFT and high-energy physics preprints. The model is designed for instruction-following scientific question answering within its specialized domain.

Key Capabilities

  • SMEFT operator reasoning: Understands and processes complex SMEFT operators.
  • EFT basis translation: Facilitates translation between different Effective Field Theory bases.
  • Physics-aware scientific dialogue: Engages in technical discussions with an understanding of particle physics concepts.
  • Literature-style technical explanation: Generates explanations in a style consistent with scientific literature.
  • Structured theoretical question answering: Provides precise answers to theoretical questions within the SMEFT domain.

Important Considerations

  • Domain-locked: This model is intentionally specialized and not suitable for general-purpose tasks.
  • Potential for hallucination: May occasionally hallucinate operator identities.
  • Uneven coverage: Due to 3620 training examples, coverage of the SMEFT operator space might be uneven, potentially leading to less reliable answers for rare operators or non-Warsaw bases.

Usage Note

When using tokenizer.apply_chat_template(...) with this model, ensure enable_thinking=False to prevent conflicts with the model's custom structured prompt format for reasoning, as it was not tuned with Qwen3's native thinking blocks.