---
license: cc-by-sa-4.0
datasets:
- Flexan/Actionier-English-1k
language:
- en
base_model:
- tiiuae/falcon-7b-instruct
pipeline_tag: text-generation
---
> [!IMPORTANT] **Important**  
> Actionier is a work in progress and is likely to be released in 2026.

# Model Card for Actionier (7B)

Actionier is a large language model that is able to solve advanced tasks by planning and using external tools.

## Model Details

### Model Description

Actionier is an instruct LLM trained to divide tasks into subtasks efficiently, solve them step by step, call tools along the way,
and stay communicative with the user. It sends multiple responses and can use multiple tools after eachother.
The model was LoRA fine-tuned with [Falcon-7B](https://huggingface.co/tiiuae/falcon-7b) as base model.

### Chat Format

Actionier uses the ChatML format, e.g.:
```text
<|im_start|>system
System message<|im_end|>
<|im_start|>user
User prompt<|im_end|>
<|im_start|>assistant
Assistant response<|im_end|>
```

### Model Usage

Details on usage will follow at a later date.