Llama Xlam 2 8b Fc R Gguf

🚀 xLAM-2 Model Family

The xLAM-2 Model Family is an advanced set of language models. It translates user intentions into executable actions, enhancing decision - making. These models are the "brains of AI agents", autonomously planning and executing tasks to reach specific goals, which is extremely useful for automating workflows in various domains. This model release is for research purposes only.

[Paper] | [Homepage] | [Dataset] | [Github]

🚀 Quick Start

This repo provides the GGUF format for the Llama - xLAM - 2 - 8b - fc - r model. Here's a link to the original model [Llama - xLAM - 2 - 8b - fc - r](https://huggingface.co/Salesforce/Llama - xLAM - 2 - 8b - fc - r).

✨ Features

• Advanced Functionality: [Large Action Models (LAMs)](https://blog.salesforceairesearch.com/large - action - models/) are designed to enhance decision - making by translating user intentions into executable actions. They serve as the "brains of AI agents", autonomously planning and executing tasks to achieve specific goals.
• Significant Leap: The new xLAM - 2 series, built on advanced data synthesis, processing, and training pipelines, shows great improvement in multi - turn conversation and tool usage.
• High - Quality Training: Trained using the novel APIGen - MT framework, which generates high - quality training data through simulated agent - human interactions.
• State - of - the - Art Performance: Achieves state - of - the - art performance on BFCL and Ï„ - bench benchmarks, outperforming frontier models like GPT - 4o and Claude 3.5.
• Refined Integration: The chat template and vLLM integration have been refined, making it easier to build advanced AI agents.

Comparative performance of larger xLAM - 2 - fc - r models (8B - 70B, trained with APIGen - MT data) against state - of - the - art baselines on function - calling (BFCL v3, as of date 04/02/2025) and agentic (Ï„ - bench) capabilities.

📦 Installation

How to Download GGUF Files

1. Install Hugging Face CLI:

pip install huggingface - hub

2. Login to Hugging Face:

huggingface - cli login

3. Download the GGUF model:

huggingface - cli download Salesforce/Llama - xLAM - 2 - 8b - fc - r - gguf Llama - xLAM - 2 - 8b - fc - r - gguf --local - dir. --local - dir - use - symlinks False

💻 Usage Examples

Basic Usage

Command Line

1. Install llama.cpp framework from the source here
2. Run the inference task as shown below. For configuration of generation - related parameters, refer to llama.cpp documentation

llama - cli -m [PATH - TO - LOCAL - GGUF]

Advanced Usage

Python Framework

1. Install [llama - cpp - python](https://github.com/abetlen/llama - cpp - python)

pip install llama - cpp - python

2. Use the high - level API for inference:

from llama_cpp import Llama
llm = Llama(
      model_path="[PATH - TO - MODEL]"
)
output = llm.create_chat_completion(
      messages = [
        {
          "role": "system",
          "content": "You are a helpful assistant that can use tools. You are developed by Salesforce xLAM team."

        },
        {
          "role": "user",
          "content": "Extract Jason is 25 years old"
        }
      ],
      tools=[{
        "type": "function",
        "function": {
          "name": "UserDetail",
          "parameters": {
            "type": "object",
            "title": "UserDetail",
            "properties": {
              "name": {
                "title": "Name",
                "type": "string"
              },
              "age": {
                "title": "Age",
                "type": "integer"
              }
            },
            "required": [ "name", "age" ]
          }
        }
      }],
      tool_choice={
        "type": "function",
        "function": {
          "name": "UserDetail"
        }
      }
)
print(output['choices'][0]['message'])

📚 Documentation

Model Series

[xLAM](https://huggingface.co/collections/Salesforce/xlam - models - 65f00e2a0a63bbcd1c2dade4) series are significantly better at many things including general tasks and function calling. For the same number of parameters, the model have been fine - tuned across a wide range of agent tasks and scenarios, all while preserving the capabilities of the original model.

*Note: The default context length for Qwen - 2.5 - based models is 32k, but you can use techniques like YaRN (Yet Another Recursive Network) to achieve a maximum 128k context length. Please refer to [here](https://huggingface.co/Qwen/Qwen2.5 - 32B - Instruct#processing - long - texts) for more details.

You can also explore our previous xLAM series [here](https://huggingface.co/collections/Salesforce/xlam - models - 65f00e2a0a63bbcd1c2dade4).

The -fc suffix indicates that the models are fine - tuned for function calling tasks, while the -r suffix signifies a research release.

✅ All models are fully compatible with vLLM and Transformers - based inference frameworks.

Prompt Template

The GGUF models use the following prompt template:

<|begin_of_text|><|start_header_id|>system<|end_header_id|>

{TASK_INSTRUCTION}
You have access to a set of tools. When using tools, make calls in a single JSON array: 

[{"name": "tool_call_name", "arguments": {"arg1": "value1", "arg2": "value2"}}, ... (additional parallel tool calls as needed)]

If no tool is suitable, state that explicitly. If the user's input lacks required parameters, ask for clarification. Do not interpret or respond until tool results are returned. Once they are available, process them or make additional calls if needed. For tasks that don't require tools, such as casual conversation or general advice, respond directly in plain text. The available tools are:

{AVAILABLE_TOOLS}

<|eot_id|><|start_header_id|>user<|end_header_id|>

{USER_QUERY}<|eot_id|><|start_header_id|>assistant<|end_header_id|>

{ASSISTANT_QUERY}<|eot_id|><|start_header_id|>user<|end_header_id|>

{USER_QUERY}<|eot_id|><|start_header_id|>assistant<|end_header_id|>

Benchmark Results

Berkeley Function - Calling Leaderboard (BFCL v3)

Performance comparison of different models on [BFCL leaderboard](https://gorilla.cs.berkeley.edu/leaderboard.html). The rank is based on the overall accuracy, which is a weighted average of different evaluation categories. "FC" stands for function - calling mode in contrast to using a customized "prompt" to extract the function calls.

Ï„ - bench Benchmark

Success Rate (pass@1) on Ï„ - bench benchmark averaged across at least 5 trials. Our xLAM - 2 - 70b - fc - r model achieves an overall success rate of 56.2% on Ï„ - bench, significantly outperforming the base Llama 3.1 70B Instruct model (38.2%) and other open - source models like DeepSeek v3 (40.6%). Notably, our best model even outperforms proprietary models such as GPT - 4o (52.9%) and approaches the performance of more recent models like Claude 3.5 Sonnet (new) (60.1%).

Pass^k curves measuring the probability that all 5 independent trials succeed for a given task, averaged across all tasks for Ï„ - retail (left) and Ï„ - airline (right) domains. Higher values indicate better consistency of the models.

Ethical Considerations

This release is for research purposes only in support of an academic paper. Our models, datasets, and code are not specifically designed or evaluated for all downstream purposes. We strongly recommend users evaluate and address potential concerns related to accuracy, safety, and fairness before deploying this model. We encourage users to consider the common limitations of AI, comply with applicable laws, and leverage best practices when selecting use cases, particularly for high - risk scenarios where errors or misuse could significantly impact people's lives, rights, or safety. For further guidance on use cases, refer to our AUP and AI AUP.

Model Licenses

For all Llama relevant models, please also follow the corresponding Llama license and terms. Meta Llama 3 is licensed under the Meta Llama 3 Community License, Copyright © Meta Platforms, Inc. All Rights Reserved.

🔧 Technical Details

The new xLAM - 2 series is built on advanced data synthesis, processing, and training pipelines. It is trained using the novel APIGen - MT framework, which generates high - quality training data through simulated agent - human interactions.

📄 License

The model release is under the cc - by - nc - 4.0 license. For all Llama relevant models, please also follow the corresponding Llama license and terms. Meta Llama 3 is licensed under the Meta Llama 3 Community License, Copyright © Meta Platforms, Inc. All Rights Reserved.

📚 Citation

If you use our model or dataset in your work, please cite our paper:

@article{prabhakar2025apigenmt,
  title={APIGen-MT: Agentic Pipeline for Multi-Turn Data Generation via Simulated Agent-Human Interplay},
  author={Prabhakar, Akshara and Liu, Zuxin and Yao, Weiran and Zhang, Jianguo and Zhu, Ming and Wang, Shiyu and Liu, Zhiwei and Awalgaonkar, Tulika and Chen, Haolin and Hoang, Thai and Niebles, Juan Carlos and Heinecke, Shelby and Wang, Huan and Savarese, Silvio and Xiong, Caiming},
  journal={arXiv preprint arXiv:2504.03601},
  year={2025}
}

Additionally, please check our other amazing works regarding the xLAM series and consider citing them as well:

@article{zhang2025actionstudio,
  title={ActionStudio: A Lightweight Framework for Data and Training of Action Models},
  author={Zhang, Jianguo and Hoang, Thai and Zhu, Ming and Liu, Zuxin and Wang, Shiyu and Awalgaonkar, Tulika and Prabhakar, Akshara and Chen, Haolin and Yao, Weiran and Liu, Zhiwei and others},
  journal={arXiv preprint arXiv:2503.22673},
  year={2025}
}

@article{zhang2024xlam,
  title={xLAM: A Family of Large Action Models to Empower AI Agent Systems},
  author={Zhang, Jianguo and Lan, Tian and Zhu, Ming and Liu, Zuxin and Hoang, Thai and Kokane, Shirley and Yao, Weiran and Tan, Juntao and Prabhakar, Akshara and Chen, Haolin and others},
  journal={arXiv preprint arXiv:2409.03215},
  year={2024}
}

@article{liu2024apigen,
  title={Apigen: Automated pipeline for generating verifiable and diverse function-calling datasets},
  author={Liu, Zuxin and Hoang, Thai and Zhang, Jianguo and Zhu, Ming and Lan, Tian and Tan, Juntao and Yao, Weiran and Liu, Zhiwei and Feng, Yihao and RN, Rithesh and others},
  journal={Advances in Neural Information Processing Systems},
  volume={37},
  pages={54463--54482},
  year={2024}
}

@article{zhang2024agentohana,
  title={AgentOhana: Design Unified Data and Training Pipeline for Effective Agent Learning},
  author={Zhang, Jianguo and Lan, Tian and Murthy, Rithesh and Liu, Zhiwei and Yao, Weiran and Tan, Juntao and Hoang, Thai and Yang, Liangwei and Feng, Yihao and Liu, Zuxin and others},
  journal={arXiv preprint arXiv:2402.15506},
  year={2024}
}