Instructions to use md-nishat-008/TigerCoder-1B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use md-nishat-008/TigerCoder-1B with Transformers:

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

pipe = pipeline("text-generation", model="md-nishat-008/TigerCoder-1B")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("md-nishat-008/TigerCoder-1B")
model = AutoModelForCausalLM.from_pretrained("md-nishat-008/TigerCoder-1B", 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=256)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use md-nishat-008/TigerCoder-1B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "md-nishat-008/TigerCoder-1B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "md-nishat-008/TigerCoder-1B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/md-nishat-008/TigerCoder-1B

SGLang

How to use md-nishat-008/TigerCoder-1B 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 "md-nishat-008/TigerCoder-1B" \
    --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": "md-nishat-008/TigerCoder-1B",
		"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 "md-nishat-008/TigerCoder-1B" \
        --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": "md-nishat-008/TigerCoder-1B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use md-nishat-008/TigerCoder-1B with Docker Model Runner:
```
docker model run hf.co/md-nishat-008/TigerCoder-1B
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

🐯 TigerCoder-1B

A compact Bangla code-generation LLM from the TigerCoder family (LREC 2026)

Nishat Raihan, Antonios Anastasopoulos, Marcos Zampieri · George Mason University

The current TigerCoder-1B scores 2–4% higher than the TigerCoder-1B results in the paper on the benchmarks below.

TigerCoder-9B is deprecated. The larger TigerCoder model is now TigerCoder-12B, which scores 5–7% higher than the 9B model.

Model summary

TigerCoder is a family of code LLMs for Bangla. TigerCoder-1B is the small member: you describe a programming task in Bangla and it writes the code. It was fine-tuned from TigerLLM-1B-it on the 300K instruction-code pairs of Bangla-Code-Instruct, and it is light enough for a laptop GPU or a CPU. It introduces itself as TigerCoder and knows who built it.


Architecture	`Gemma3ForCausalLM` (Gemma 3, 1B)
Parameters	1.0B
Precision	BF16, one `model.safetensors` file (about 1.9 GiB)
Context length	32,768 tokens (from `config.json`)
Base model	TigerLLM-1B-it
Task	Code generation from Bangla prompts
Programming languages	Python, C++, Java, JavaScript
Fine-tuning data	Bangla-Code-Instruct, 300K Bangla instruction-code pairs
License	CC BY 4.0

Benchmarks

Pass@K on Python tasks with Bangla prompts, from the TigerCoder paper (LREC 2026).

Model	mHumanEval P@1	mHumanEval P@10	mHumanEval P@100	MBPP P@1	MBPP P@10	MBPP P@100
GPT-3.5	0.56	0.56	0.59	0.60	0.62	0.62
Gemini-Flash 2.5	0.58	0.61	0.62	0.62	0.62	0.70
Gemma-3 (27B)	0.64	0.65	0.69	0.69	0.70	0.70
TigerLLM (9B) (deprecated)	0.63	0.69	0.72	0.61	0.68	0.73
TigerCoder-1B, paper checkpoint	0.69	0.73	0.77	0.74	0.74	0.81
TigerCoder-1B, current (this model)	+2–4%	+2–4%	+2–4%	+2–4%	+2–4%	+2–4%
TigerCoder-9B (deprecated)	0.75	0.80	0.84	0.82	0.84	0.91
TigerCoder-12B	+5–7%	+5–7%	+5–7%	+5–7%	+5–7%	+5–7%

The numeric rows are from the paper. This model scores 2–4% higher than the paper's TigerCoder-1B, and TigerCoder-12B scores 5–7% higher than TigerCoder-9B.

The paper also tested TigerCoder-1B in other programming languages (Pass@1, Bangla prompts):

Language	mHumanEval P@1	MBPP P@1
Python	0.69	0.74
C++	0.64	0.66
Java	0.58	0.61
JavaScript	0.53	0.55

Quick start

pip install "torch>=2.4" "transformers>=4.56" accelerate sentencepiece

The model needs about 2 GB of GPU memory in BF16, and it also runs on CPU or Apple Silicon (device_map="mps").

import torch
from transformers import AutoModelForCausalLM, AutoTokenizer

MODEL_ID = "md-nishat-008/TigerCoder-1B"

tokenizer = AutoTokenizer.from_pretrained(MODEL_ID)
model = AutoModelForCausalLM.from_pretrained(
    MODEL_ID,
    dtype=torch.bfloat16,
    device_map="auto",
).eval()

# "Write a Python function that computes the factorial of a number."
messages = [{"role": "user", "content": "পাইথনে একটি ফাংশন লিখুন যা একটি সংখ্যার ফ্যাক্টরিয়াল গণনা করে।"}]

inputs = tokenizer.apply_chat_template(
    messages,
    add_generation_prompt=True,
    return_dict=True,
    return_tensors="pt",
).to(model.device)

with torch.inference_mode():
    output_ids = model.generate(**inputs, max_new_tokens=512, do_sample=False, repetition_penalty=1.05)

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

Prompting tips

Build prompts with tokenizer.apply_chat_template(...). Do not paste special chat tokens by hand.
Name the programming language, the function name and signature, and the expected input and output, for example: পাইথনে একটি ফাংশন second_largest(nums) লিখুন যা একটি সংখ্যার তালিকা থেকে দ্বিতীয় বৃহত্তম ভিন্ন সংখ্যাটি ফেরত দেয়।
Use do_sample=False for Pass@1-style answers. For several candidate solutions (Pass@10, Pass@100), sample with do_sample=True, temperature=0.7, top_p=0.95.
A 1B model handles short, self-contained functions best. For longer programs or tricky edge cases, use TigerCoder-12B.

Training

The current checkpoint is the TigerCoder-1B code fine-tune with a short identity fine-tune on top.

Identity fine-tune	Value
Data	2,000 TigerCoder identity examples (Bangla and English) mixed with 6,000 Bangla-Code-Instruct examples, so coding skill is kept
Method	LoRA (rank 16, alpha 32) on all attention and MLP projections, merged into the weights
Epochs / learning rate	2 / 1e-4, cosine schedule
Check	loss on 500 held-out Bangla-Code-Instruct examples went from 1.030 to 1.029

Code fine-tuning setup the paper reports for TigerCoder-1B

Hyperparameter	Value
Base model	TigerLLM-1B-it
Training data	Bangla-Code-Instruct (300K examples)
Max sequence length	2048
Batch size (train / eval)	16
Gradient accumulation steps	4
Epochs	3
Learning rate	1 × 10⁻⁵
Weight decay	0.02
Warm-up	10% of steps
Optimizer	AdamW
LR scheduler	Cosine
Precision	BF16
Hardware	NVIDIA A100 (40 GB)

Training data

Bangla-Code-Instruct has three 100K subsets:

Subset	Size	How it was built	Prompt origin	Code origin
SI (Self-Instruct)	100K	5,000 expert seeds expanded with GPT-4o	Semi-natural	Synthetic
Syn (Synthetic)	100K	Generated with GPT-4o and Claude 3.5 Sonnet	Synthetic	Synthetic
TE (Translated)	100K	Evol-Instruct machine-translated with NLLB-200	Translated	Natural

All SI and Syn code passed a syntax check (ast.parse) and an execution test (Python 3.13.0, 10 s timeout, 16 GB memory).

Findings from the paper

Bangla prompts hurt code LLMs. Most models score 20 to 50+ percentage points lower on Bangla coding prompts than on the same prompts in English.
Translating the prompt to English does not fix it. Machine translation garbles programming terms: "অক্ষর" (character) becomes "letter", "চলক" (variable) becomes "clever", and "স্ট্রিং" (string) becomes "rope".
Targeted data beats scale. TigerCoder-1B beats models 27 times its size.

Repository files

File	Contents
`model.safetensors`	Full BF16 model weights
`config.json`, `generation_config.json`	Model and generation settings
`tokenizer.json`, `tokenizer.model`, `tokenizer_config.json`, `chat_template.jinja`	Tokenizer and chat format

Limitations

TigerCoder is trained for Bangla code generation. On general language tasks, use TigerLLM-1B-it.
The training data is synthetic or machine-translated, and may carry the biases and artifacts of the models that produced it.
Evaluation covers MBPP-Bangla and mHumanEval-Bangla only. Production-scale coding tasks have not been benchmarked.
Run and review generated code before you use it.

License

CC BY 4.0. Check the Gemma 3 terms of use and the dataset licenses before redistribution or commercial use.

Citation

@inproceedings{raihan-etal-2026-tigercoder,
    title = "{T}iger{C}oder: A Novel Suite of {LLM}s for Code Generation in {B}angla",
    author = "Raihan, Nishat  and
      Anastasopoulos, Antonios  and
      Zampieri, Marcos",
    editor = "Piperidis, Stelios  and
      Bel, N{\'u}ria  and
      van den Heuvel, Henk  and
      Ide, Nancy  and
      Krek, Simon  and
      Toral, Antonio",
    booktitle = "Proceedings of the Fifteenth Language Resources and Evaluation Conference",
    month = may,
    year = "2026",
    address = "Palma de Mallorca, Spain",
    publisher = "ELRA Language Resource Association",
    url = "https://aclanthology.org/2026.lrec-1.238/",
    doi = "10.63317/5nampb63np3m",
    pages = "3044--3054"
}

Related work

@inproceedings{raihan-zampieri-2025-tigerllm,
    title = "{T}iger{LLM} - A Family of {B}angla Large Language Models",
    author = "Raihan, Nishat  and
      Zampieri, Marcos",
    editor = "Che, Wanxiang  and
      Nabende, Joyce  and
      Shutova, Ekaterina  and
      Pilehvar, Mohammad Taher",
    booktitle = "Proceedings of the 63rd Annual Meeting of the {A}ssociation for {C}omputational {L}inguistics (Volume 2: Short Papers)",
    month = jul,
    year = "2025",
    address = "Vienna, Austria",
    publisher = "Association for Computational Linguistics",
    url = "https://aclanthology.org/2025.acl-short.69/",
    doi = "10.18653/v1/2025.acl-short.69",
    pages = "887--896",
    ISBN = "979-8-89176-252-7"
}

@inproceedings{raihan-etal-2025-mhumaneval,
    title = "m{H}uman{E}val - A Multilingual Benchmark to Evaluate Large Language Models for Code Generation",
    author = "Raihan, Nishat  and
      Anastasopoulos, Antonios  and
      Zampieri, Marcos",
    editor = "Chiruzzo, Luis  and
      Ritter, Alan  and
      Wang, Lu",
    booktitle = "Proceedings of the 2025 Conference of the Nations of the {A}mericas Chapter of the {A}ssociation for {C}omputational {L}inguistics: Human Language Technologies (Volume 1: Long Papers)",
    month = apr,
    year = "2025",
    address = "Albuquerque, New Mexico",
    publisher = "Association for Computational Linguistics",
    url = "https://aclanthology.org/2025.naacl-long.570/",
    doi = "10.18653/v1/2025.naacl-long.570",
    pages = "11432--11461",
    ISBN = "979-8-89176-189-6"
}

Questions: mraihan2@gmu.edu

Downloads last month: 1,398

Safetensors

Model size

1.0B params

Tensor type

BF16

Model tree for md-nishat-008/TigerCoder-1B

Base model

md-nishat-008/TigerLLM-1B-it

Finetuned

(2)

this model

Quantizations

1 model

md-nishat-008
/

TigerCoder-1B