Instructions to use Navyaforaa/LitGram-1.5B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Navyaforaa/LitGram-1.5B with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="Navyaforaa/LitGram-1.5B")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Navyaforaa/LitGram-1.5B")
model = AutoModelForCausalLM.from_pretrained("Navyaforaa/LitGram-1.5B", device_map="auto")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use Navyaforaa/LitGram-1.5B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Navyaforaa/LitGram-1.5B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Navyaforaa/LitGram-1.5B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Navyaforaa/LitGram-1.5B

SGLang

How to use Navyaforaa/LitGram-1.5B with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "Navyaforaa/LitGram-1.5B" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Navyaforaa/LitGram-1.5B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "Navyaforaa/LitGram-1.5B" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Navyaforaa/LitGram-1.5B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Unsloth Studio

How to use Navyaforaa/LitGram-1.5B 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 Navyaforaa/LitGram-1.5B 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 Navyaforaa/LitGram-1.5B to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for Navyaforaa/LitGram-1.5B to start chatting

Load model with FastModel

pip install unsloth
from unsloth import FastModel
model, tokenizer = FastModel.from_pretrained(
    model_name="Navyaforaa/LitGram-1.5B",
    max_seq_length=2048,
)

Docker Model Runner
How to use Navyaforaa/LitGram-1.5B with Docker Model Runner:
```
docker model run hf.co/Navyaforaa/LitGram-1.5B
```

📚 LitGram-1.5B: Domain-Adapted LLM for English Literature & Grammar

Developed by: Navyaforaa • Model Size: 1.54 Billion Parameters • Base Architecture: Qwen2.5-1.5B-Instruct
Training Framework: Unsloth AI + PyTorch + TRL • License: Apache 2.0

📖 Model Overview

LitGram-1.5B is a specialized, domain-adapted language model tailored for English literature scholars, students, UGC NET English aspirants, literary critics, and grammar/syntax researchers.

While standard small language models often struggle with complex literary criticism—frequently hallucinating generic lists of buzzwords or failing on precise metrical scansion—LitGram has been fine-tuned on an extensive, multi-stage curated dataset designed for text-grounded literary hermeneutics, line-by-line prosody, and rigorous syntactic parsing.

🎯 Key Capabilities & Core Domains

1. Applied Literary Theory & Hermeneutics

Frameworks Covered: New Historicism, Cultural Materialism, Marxism, Feminism, Queer Theory, Psychoanalysis (Freudian & Lacanian), Postcolonial Theory, Ecocriticism, Deconstruction, Reader-Response Theory, and Cultural Studies.
Defensible Critiques: Answers follow a structured 3-part academic methodology:
1. Core Theoretical Premise
2. Textual and Historical Evidentiary Grounding
3. Critical Limitations and Alternative Readings
Deep Author Grounding: Specialized corpora for John Milton (Paradise Lost, Samson Agonistes, Lycidas, Areopagitica), William Shakespeare, Geoffrey Chaucer, Jane Austen, Charlotte & Emily Brontë, Charles Dickens, Virginia Woolf, James Joyce, and Salman Rushdie.

2. Poetics, Metrical Scansion & Classical Rhetoric

Line-by-Line Scansion: Accurately identifies syllable stress (x /), metric feet (iamb, trochee, anapest, dactyl, spondee, pyrrhic), and metric substitutions (e.g. initial trochees, feminine endings, caesuras).
Poetic Forms: Sonnets (Petrarchan, Shakespearean, Spenserian), Villanelles, Terza Rima, Ottava Rima, Spenserian Stanzas, Rhyme Royal, and Heroic Couplets.
Rhetorical Figures: Chiasmus vs. Antimetabole, Metonymy vs. Synecdoche, Litotes, Hypallage, Zeugma, Symploce, and Asyndeton.

3. Advanced English Grammar & Generative Syntax

Syntactic Parsing: Clause hierarchy, X-Bar phrase structure breakdown, and constituent analysis.
Prescriptive & Descriptive Mechanics: Correction of dangling, squinting, and misplaced modifiers; Mandative and counterfactual irrealis were-subjunctive; non-finite verbals (gerund vs. participle vs. infinitive); pronoun case in comparative ellipses; and punctuation conventions.

4. UGC NET English Literature & Research Methodology

Specialized Units: Indian Writing in English (Raja Rao, Mulk Raj Anand, R.K. Narayan), Dalit Literature aesthetics (Limbale, Bama, Valmiki), English Language Teaching (ELT & Krashen's Hypotheses), British Cultural Studies (Birmingham CCCS, Stuart Hall, Raymond Williams), and MLA 9th Edition research standards.

5. Project Gutenberg Integration

Grounded in canonical English public-domain texts across poetry, drama, essays, and Victorian/Romantic fiction.

💻 How to Use LitGram in Python

Using Hugging Face Transformers

import torch
from transformers import AutoModelForCausalLM, AutoTokenizer

model_name = "Navyaforaa/LitGram-1.5B"

tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(
    model_name,
    torch_dtype=torch.float16,
    device_map="auto"
)

messages = [
    {
        "role": "system",
        "content": "You are LitGram, a specialized model for English literature and English grammar. Give accurate, text-grounded, clearly explained answers."
    },
    {
        "role": "user",
        "content": "Scan the metre of Shakespeare's Sonnet 18: 'Shall I compare thee to a summer's day?'"
    }
]

prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
inputs = tokenizer(prompt, return_tensors="pt").to("cuda")

outputs = model.generate(
    **inputs,
    max_new_tokens=512,
    temperature=0.3,
    repetition_penalty=1.15
)

print(tokenizer.decode(outputs[0][inputs.input_ids.shape[1]:], skip_special_tokens=True))

Downloads last month: 640

Safetensors

Model size

2B params

Tensor type

BF16

Model tree for Navyaforaa/LitGram-1.5B

Base model

Qwen/Qwen2.5-1.5B

Finetuned

Qwen/Qwen2.5-1.5B-Instruct

Finetuned

(1773)

this model

Finetunes

2 models