Tmax-2B-GGUF

Direct GGUF Quantizations of Tmax-2b

This repository provides GGUF quantized models for allenai/tmax-2b.

Tmax-2b is a 2 billion parameter terminal-agent model developed by AllenAI and collaborators. Built on top of the Qwen3.5-2B architecture and further trained using reinforcement learning for terminal-based tasks, it is designed to perform complex command-line and software engineering workflows while maintaining strong general-purpose reasoning capabilities. Note that the vision head was removed during training, making this a pure text-based model. These GGUF versions are optimized for efficient CPU and GPU inference using llama.cpp and compatible tools.

This release includes various quantization levels (e.g., Q2_K, Q3_K_M, Q4_K_M, Q5_K_M, Q6_K, Q8_0) to suit different hardware capabilities and performance requirements.

Table of Contents 📝

1. ▶ Usage
2. 📃 License
3. 🙏 Acknowledgements

pip install "huggingface_hub[cli]"
huggingface-cli download samuelchristlie/tmax-2b-gguf --local-dir ./tmax-2b-gguf

You can also download directly from this page

2. Inference

To use these GGUF files, you'll need a compatible inference engine like llama.cpp or clients built on top of it (e.g., Ollama, LM Studio, KoboldCpp, text-generation-webui with llama.cpp backend).

📃 License

This model is a GGUF conversion of the original allenai/tmax-2b model. The original model is licensed under the Apache 2.0 License, and this derivative work adheres to the terms of that license. Please review the original license for full details.

🙏 Acknowledgements

• Allen Institute for AI (AI2) and collaborators for developing and open-sourcing the tmax-2b model:
  • https://huggingface.co/allenai/tmax-2b
• Qwen Team for the Qwen3.5-2B base model that powers tmax-2b:
  • https://huggingface.co/Qwen/Qwen3.5-2B
• The llama.cpp project and its contributors for the GGUF format and the incredible tooling that makes local LLM inference accessible.
  • https://github.com/ggml-org/llama.cpp
• city96:
  • https://huggingface.co/city96