---
license: apache-2.0
library_name: transformers
pipeline_tag: text-generation
language:
- ru
- en
tags:
- catrex
- llama
- gguf
- text-generation
datasets:
- HuggingFaceFW/fineweb-edu
- HuggingFaceH4/ultrachat_200k
- IlyaGusev/saiga_scored
- d0rj/OpenOrca-ru
---

# Catrex 1.0

Bilingual (RU/EN) chat model on **65M parameters**, trained **from scratch**.

## Results

| Metric | Value |
|---|---|
| val loss | **4.8311** |
| perplexity | 125.3 |
| Parameters | 64.82M |
| Context | 512 tokens |
| Training steps | 300 |

## Run in LM Studio
1. Download `catrex-1.0-f16.gguf`
2. Put it in `~/.lmstudio/models/Catniti/catrex-1.0/`
3. Load it in LM Studio (chat template is already inside the file)

## Run via llama.cpp
```bash
llama-cli -hf Catniti/catrex-1.0:f16 -p "Hi!"
```

## Python (transformers)
```python
from transformers import AutoModelForCausalLM, AutoTokenizer
tok = AutoTokenizer.from_pretrained("Catniti/catrex-1.0")
model = AutoModelForCausalLM.from_pretrained("Catniti/catrex-1.0")
msgs = [{"role": "user", "content": "Hi!"}]
ids = tok.apply_chat_template(msgs, add_generation_prompt=True, return_tensors="pt")
out = model.generate(ids, max_new_tokens=100, temperature=0.8, do_sample=True)
print(tok.decode(out[0][ids.shape[1]:], skip_special_tokens=True))
```

## Prompt format
ChatML:
```
<|im_start|>user
Hi!<|im_end|>
<|im_start|>assistant
```