Instructions to use pszemraj/flan-t5-xl-grammar-synthesis with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use pszemraj/flan-t5-xl-grammar-synthesis with Transformers:

# Load model directly
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM

tokenizer = AutoTokenizer.from_pretrained("pszemraj/flan-t5-xl-grammar-synthesis")
model = AutoModelForSeq2SeqLM.from_pretrained("pszemraj/flan-t5-xl-grammar-synthesis")

llama-cpp-python

How to use pszemraj/flan-t5-xl-grammar-synthesis with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="pszemraj/flan-t5-xl-grammar-synthesis",
	filename="flan-t5-xl-grammar-synthesis-q6_k.gguf",
)

output = llm(
	"Once upon a time,",
	max_tokens=512,
	echo=True
)
print(output)

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use pszemraj/flan-t5-xl-grammar-synthesis with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf pszemraj/flan-t5-xl-grammar-synthesis:Q6_K
# Run inference directly in the terminal:
llama-cli -hf pszemraj/flan-t5-xl-grammar-synthesis:Q6_K

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf pszemraj/flan-t5-xl-grammar-synthesis:Q6_K
# Run inference directly in the terminal:
llama-cli -hf pszemraj/flan-t5-xl-grammar-synthesis:Q6_K

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf pszemraj/flan-t5-xl-grammar-synthesis:Q6_K
# Run inference directly in the terminal:
./llama-cli -hf pszemraj/flan-t5-xl-grammar-synthesis:Q6_K

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf pszemraj/flan-t5-xl-grammar-synthesis:Q6_K
# Run inference directly in the terminal:
./build/bin/llama-cli -hf pszemraj/flan-t5-xl-grammar-synthesis:Q6_K

Use Docker

docker model run hf.co/pszemraj/flan-t5-xl-grammar-synthesis:Q6_K

LM Studio
Jan
Ollama
How to use pszemraj/flan-t5-xl-grammar-synthesis with Ollama:
```
ollama run hf.co/pszemraj/flan-t5-xl-grammar-synthesis:Q6_K
```

Unsloth Studio

How to use pszemraj/flan-t5-xl-grammar-synthesis with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for pszemraj/flan-t5-xl-grammar-synthesis to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for pszemraj/flan-t5-xl-grammar-synthesis to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for pszemraj/flan-t5-xl-grammar-synthesis to start chatting

Docker Model Runner
How to use pszemraj/flan-t5-xl-grammar-synthesis with Docker Model Runner:
```
docker model run hf.co/pszemraj/flan-t5-xl-grammar-synthesis:Q6_K
```

Lemonade

How to use pszemraj/flan-t5-xl-grammar-synthesis with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull pszemraj/flan-t5-xl-grammar-synthesis:Q6_K

Run and chat with the model

lemonade run user.flan-t5-xl-grammar-synthesis-Q6_K

List all available models

lemonade list

grammar-synthesis: flan-t5-xl

This model is a fine-tuned version of google/flan-t5-xl on an extended version of the JFLEG dataset.

here is a custom class wrapper that makes using this with bitsandbytes easier
the API can be slow due to model size, try the notebook!

Model description

The intent is to create a text2text language model that successfully performs "single-shot grammar correction" on a potentially grammatically incorrect text that could have many errors with the important qualifier that it does not semantically change text/information that IS grammatically correct..

Compare some of the more severe error examples on other grammar correction models to see the difference :)

Limitations

Data set: cc-by-nc-sa-4.0
Model: apache-2.0
currently a work in progress! While probably useful for "single-shot grammar correction" in many cases, check the output for correctness, ok?.

Training procedure

Training hyperparameters

Session One

TODO: add this. It was a single epoch at higher LR

Session Two

The following hyperparameters were used during training:

learning_rate: 4e-05
train_batch_size: 4
eval_batch_size: 1
seed: 42
distributed_type: multi-GPU
gradient_accumulation_steps: 16
total_train_batch_size: 64
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: cosine
lr_scheduler_warmup_ratio: 0.02
num_epochs: 2.0

Downloads last month: 90

Safetensors

Model size

3B params

Tensor type

F32

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for pszemraj/flan-t5-xl-grammar-synthesis

Base model

google/flan-t5-xl

Quantized

(3)

this model

Dataset used to train pszemraj/flan-t5-xl-grammar-synthesis

Space using pszemraj/flan-t5-xl-grammar-synthesis 1

Collection including pszemraj/flan-t5-xl-grammar-synthesis

Grammar Synthesis

Collection

From nothing; grammar. • 6 items • Updated Jul 28, 2025