---
base_model: TinyLlama/TinyLlama-1.1B-Chat-v0.5
datasets:
- cerebras/SlimPajama-627B
- bigcode/starcoderdata
- OpenAssistant/oasst_top1_2023-08-25
inference: false
language:
- en
license: apache-2.0
model_creator: TinyLlama
model_name: TinyLlama-1.1B-Chat-v0.5
pipeline_tag: text-generation
quantized_by: afrideva
tags:
- gguf
- ggml
- quantized
- q2_k
- q3_k_m
- q4_k_m
- q5_k_m
- q6_k
- q8_0
---
# TinyLlama/TinyLlama-1.1B-Chat-v0.5-GGUF

Quantized GGUF model files for [TinyLlama-1.1B-Chat-v0.5](https://huggingface.co./TinyLlama/TinyLlama-1.1B-Chat-v0.5) from [TinyLlama](https://huggingface.co./TinyLlama)


| Name | Quant method | Size |
| ---- | ---- | ---- |
| [tinyllama-1.1b-chat-v0.5.q2_k.gguf](https://huggingface.co./afrideva/TinyLlama-1.1B-Chat-v0.5-GGUF/resolve/main/tinyllama-1.1b-chat-v0.5.q2_k.gguf) | q2_k | 482.15 MB  |
| [tinyllama-1.1b-chat-v0.5.q3_k_m.gguf](https://huggingface.co./afrideva/TinyLlama-1.1B-Chat-v0.5-GGUF/resolve/main/tinyllama-1.1b-chat-v0.5.q3_k_m.gguf) | q3_k_m | 549.85 MB  |
| [tinyllama-1.1b-chat-v0.5.q4_k_m.gguf](https://huggingface.co./afrideva/TinyLlama-1.1B-Chat-v0.5-GGUF/resolve/main/tinyllama-1.1b-chat-v0.5.q4_k_m.gguf) | q4_k_m | 667.82 MB  |
| [tinyllama-1.1b-chat-v0.5.q5_k_m.gguf](https://huggingface.co./afrideva/TinyLlama-1.1B-Chat-v0.5-GGUF/resolve/main/tinyllama-1.1b-chat-v0.5.q5_k_m.gguf) | q5_k_m | 782.05 MB  |
| [tinyllama-1.1b-chat-v0.5.q6_k.gguf](https://huggingface.co./afrideva/TinyLlama-1.1B-Chat-v0.5-GGUF/resolve/main/tinyllama-1.1b-chat-v0.5.q6_k.gguf) | q6_k | 903.42 MB  |
| [tinyllama-1.1b-chat-v0.5.q8_0.gguf](https://huggingface.co./afrideva/TinyLlama-1.1B-Chat-v0.5-GGUF/resolve/main/tinyllama-1.1b-chat-v0.5.q8_0.gguf) | q8_0 | 1.17 GB  |



## Original Model Card:
<div align="center">

# TinyLlama-1.1B
</div>

https://github.com/jzhang38/TinyLlama

The TinyLlama project aims to **pretrain** a **1.1B Llama model on 3 trillion tokens**. With some proper optimization, we can achieve this within a span of "just" 90 days using 16 A100-40G GPUs 🚀🚀. The training has started on 2023-09-01. 


We adopted exactly the same architecture and tokenizer as Llama 2. This means TinyLlama can be plugged and played in many open-source projects built upon Llama. Besides, TinyLlama is compact with only 1.1B parameters. This compactness allows it to cater to a multitude of applications demanding a restricted computation and memory footprint.

#### This Model
This is the chat model finetuned on top of [TinyLlama/TinyLlama-1.1B-intermediate-step-955k-2T](https://huggingface.co./TinyLlama/TinyLlama-1.1B-intermediate-step-955k-token-2T). 
The dataset used is [OpenAssistant/oasst_top1_2023-08-25](https://huggingface.co./datasets/OpenAssistant/oasst_top1_2023-08-25) following the [chatml](https://github.com/openai/openai-python/blob/main/chatml.md) format.
#### How to use
You will need the transformers>=4.31
Do check the [TinyLlama](https://github.com/jzhang38/TinyLlama) github page for more information.
```
from transformers import AutoTokenizer
import transformers 
import torch
model = "PY007/TinyLlama-1.1B-Chat-v0.5"
tokenizer = AutoTokenizer.from_pretrained(model)
pipeline = transformers.pipeline(
    "text-generation",
    model=model,
    torch_dtype=torch.float16,
    device_map="auto",
)

CHAT_EOS_TOKEN_ID = 32002

prompt = "How to get in a good university?"
formatted_prompt = (
    f"<|im_start|>user\n{prompt}<|im_end|>\n<|im_start|>assistant\n"
)


sequences = pipeline(
    formatted_prompt,
    do_sample=True,
    top_k=50,
    top_p = 0.9,
    num_return_sequences=1,
    repetition_penalty=1.1,
    max_new_tokens=1024,
    eos_token_id=CHAT_EOS_TOKEN_ID,
)

for seq in sequences:
    print(f"Result: {seq['generated_text']}")
```