|
--- |
|
license: apache-2.0 |
|
language: |
|
- en |
|
- sr |
|
- hr |
|
- bs |
|
datasets: |
|
- teknium/OpenHermes-2.5 |
|
- WizardLM/WizardLM_evol_instruct_V2_196k |
|
- draganjovanovich/airoboros-3.0-serbian |
|
- Open-Orca/SlimOrca |
|
- draganjovanovich/prodigy-instruct-reason-function-calling |
|
- iskonai/chatislav-instruct |
|
- mlabonne/FineTome-100k |
|
base_model: |
|
- draganjovanovich/prodigy-sm-base-v0.1 |
|
--- |
|
# Prodigy SM Instruct v0.1 |
|
|
|
<img src="https://cdn-uploads.huggingface.co/production/uploads/617bbeec14572ebe9e6ea83f/5sl1RHE054rNkJMA4Z0dp.png" width="70%" height="70%"> |
|
|
|
**Prodigy SM Instruct v0.1** is an instruction-tuned version of our [Prodigy SM Base v0.1](https://huggingface.co/iskonai/prodigy-sm-base-v0.1) model, optimized for instruction following in Serbian, Croatian, Bosnian and English languages. The model demonstrates exceptional capabilities in function calling and tool usage while maintaining strong performance across all supported languages. |
|
|
|
Model was shared as part of the presenatation at DSC Europe 2024 / Belgrade (BREAKOUT SESSIONS & WORKSHOPS). |
|
|
|
# 🎮 DEMO TIME! 🚀 For next couple of weeks you can test model at: [https://chat.iskon.ai](https://chat.iskon.ai) |
|
or you can download it and use **SGLang**(recomended as it have superior structured output sampling), vLLM, HF text-generation-inference etc. |
|
proper model card coming soon when we upload final checkpoint, and remove **draft** from the name :) |
|
|
|
# Training |
|
|
|
The model was instruction-tuned using a carefully curated mix of high-quality datasets in ChatML format. |
|
|
|
# Features |
|
- Precise **system prompt** following capabilities in **Serbian** (trained on really large system prompts: 500+ tokens.) |
|
- Strong instruction following capabilities in Serbian, Croatian, Bosnian and English |
|
- Enhanced function calling and tool usage abilities |
|
- Maintains the base model's strong performance in all supported languages |
|
- Uses ChatML format for conversations |
|
|
|
# Use Cases |
|
|
|
The model is particularly well-suited for: |
|
- Multilingual applications requiring Serbian, Croatian, Bosnian and English support |
|
- Applications requiring precise instruction following |
|
- Systems utilizing function calling and tool usage |
|
- Conversational AI applications |
|
- Text generation in supported languages |
|
|
|
# Limitations |
|
|
|
As with all language models: |
|
- Outputs should be verified for critical applications |
|
- Performance in languages other than Serbian, Croatian, Bosnian and English may vary |
|
- Users should be aware of general LLM limitations regarding hallucination and factual accuracy |
|
|
|
# What's Next |
|
|
|
Stay tuned for our upcoming releases: |
|
- [prodigy-sm-base (llama-3.1)]() **COMING SOON** |
|
- [prodigy-sm-instruct (llama-3.1)]() **COMING SOON** |
|
|
|
📢 We're excited to announce that [iskon.ai](https://Iskon.ai) will soon launch an API platform featuring advanced **Prodigy** series of models, advanced AI tools and much more! 🚀 |
