RichardErkhov/unsloth_-_tinyllama-gguf

Quantization made by Richard Erkhov.

Github

Discord

Request more models

tinyllama - GGUF

• Model creator: https://huggingface.co/unsloth/
• Original model: https://huggingface.co/unsloth/tinyllama/

Name	Quant method	Size
tinyllama.Q2_K.gguf	Q2_K	0.4GB
tinyllama.IQ3_XS.gguf	IQ3_XS	0.44GB
tinyllama.IQ3_S.gguf	IQ3_S	0.47GB
tinyllama.Q3_K_S.gguf	Q3_K_S	0.47GB
tinyllama.IQ3_M.gguf	IQ3_M	0.48GB
tinyllama.Q3_K.gguf	Q3_K	0.51GB
tinyllama.Q3_K_M.gguf	Q3_K_M	0.51GB
tinyllama.Q3_K_L.gguf	Q3_K_L	0.55GB
tinyllama.IQ4_XS.gguf	IQ4_XS	0.57GB
tinyllama.Q4_0.gguf	Q4_0	0.59GB
tinyllama.IQ4_NL.gguf	IQ4_NL	0.6GB
tinyllama.Q4_K_S.gguf	Q4_K_S	0.6GB
tinyllama.Q4_K.gguf	Q4_K	0.62GB
tinyllama.Q4_K_M.gguf	Q4_K_M	0.62GB
tinyllama.Q4_1.gguf	Q4_1	0.65GB
tinyllama.Q5_0.gguf	Q5_0	0.71GB
tinyllama.Q5_K_S.gguf	Q5_K_S	0.71GB
tinyllama.Q5_K.gguf	Q5_K	0.73GB
tinyllama.Q5_K_M.gguf	Q5_K_M	0.73GB
tinyllama.Q5_1.gguf	Q5_1	0.77GB
tinyllama.Q6_K.gguf	Q6_K	0.84GB
tinyllama.Q8_0.gguf	Q8_0	1.09GB

Original model description:

language: - en library_name: transformers license: apache-2.0 tags: - unsloth - transformers - tinyllama

Finetune Mistral, Gemma, Llama 2-5x faster with 70% less memory via Unsloth!

A reupload from https://huggingface.co/TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T

We have a Google Colab Tesla T4 notebook for TinyLlama with 4096 max sequence length RoPE Scaling here: https://colab.research.google.com/drive/1AZghoNBQaMDgWJpi4RbffGM1h6raLUj9?usp=sharing

✨ Finetune for Free

All notebooks are beginner friendly! Add your dataset, click "Run All", and you'll get a 2x faster finetuned model which can be exported to GGUF, vLLM or uploaded to Hugging Face.

Unsloth supports	Free Notebooks	Performance	Memory use
Gemma 7b	▶️ Start on Colab	2.4x faster	58% less
Mistral 7b	▶️ Start on Colab	2.2x faster	62% less
Llama-2 7b	▶️ Start on Colab	2.2x faster	43% less
TinyLlama	▶️ Start on Colab	3.9x faster	74% less
CodeLlama 34b A100	▶️ Start on Colab	1.9x faster	27% less
Mistral 7b 1xT4	▶️ Start on Kaggle	5x faster*	62% less
DPO - Zephyr	▶️ Start on Colab	1.9x faster	19% less

• This conversational notebook is useful for ShareGPT ChatML / Vicuna templates.
• This text completion notebook is for raw text. This DPO notebook replicates Zephyr.
• * Kaggle has 2x T4s, but we use 1. Due to overhead, 1x T4 is 5x faster.