Instructions to use ajibawa-2023/Code-Jamba-v0.1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use ajibawa-2023/Code-Jamba-v0.1 with Transformers:

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

pipe = pipeline("text-generation", model="ajibawa-2023/Code-Jamba-v0.1", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("ajibawa-2023/Code-Jamba-v0.1", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("ajibawa-2023/Code-Jamba-v0.1", trust_remote_code=True)
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

vLLM

How to use ajibawa-2023/Code-Jamba-v0.1 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "ajibawa-2023/Code-Jamba-v0.1"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ajibawa-2023/Code-Jamba-v0.1",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/ajibawa-2023/Code-Jamba-v0.1

SGLang

How to use ajibawa-2023/Code-Jamba-v0.1 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 "ajibawa-2023/Code-Jamba-v0.1" \
    --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": "ajibawa-2023/Code-Jamba-v0.1",
		"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 "ajibawa-2023/Code-Jamba-v0.1" \
        --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": "ajibawa-2023/Code-Jamba-v0.1",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use ajibawa-2023/Code-Jamba-v0.1 with Docker Model Runner:
```
docker model run hf.co/ajibawa-2023/Code-Jamba-v0.1
```

Code-Jamba-v0.1

This model is trained upon my dataset Code-290k-ShareGPT and Code-Feedback. It is finetuned on Jamba-v0.1 . It is very very good in Code generation in various languages such as Python, Java, JavaScript, GO, C++, Rust, Ruby, Sql, MySql, R, Julia, Haskell, etc.. This model will also generate detailed explanation/logic behind each code. This model uses ChatML prompt format.

Training

Entire dataset was trained on 2 x H100 94GB. For 3 epoch, training took 162 hours. Axolotl along with DeepSpeed codebase was used for training purpose. This was trained on Jamba-v0.1 by AI21Labs.

This is a qlora model. Links for quantized models will be updated very soon.

GPTQ, GGUF, AWQ & Exllama

GPTQ: TBA

GGUF: TBA

AWQ: TBA

Exllama v2: TBA

Example Prompt:

This model uses ChatML prompt format.

<|im_start|>system
You are a Helpful Assistant.<|im_end|>
<|im_start|>user
{prompt}<|im_end|>
<|im_start|>assistant

You can modify above Prompt as per your requirement.

I want to say special Thanks to the Open Source community for helping & guiding me to better understand the AI/Model development.

Thank you for your love & support.

Example Output

Coming soon!

Downloads last month: 16

Safetensors

Model size

52B params

Tensor type

BF16

ajibawa-2023
/

Code-Jamba-v0.1

Datasets used to train ajibawa-2023/Code-Jamba-v0.1