Instructions to use state-spaces/mamba-790m-hf with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use state-spaces/mamba-790m-hf with Transformers:

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

pipe = pipeline("text-generation", model="state-spaces/mamba-790m-hf")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("state-spaces/mamba-790m-hf")
model = AutoModelForCausalLM.from_pretrained("state-spaces/mamba-790m-hf")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use state-spaces/mamba-790m-hf with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "state-spaces/mamba-790m-hf"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "state-spaces/mamba-790m-hf",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/state-spaces/mamba-790m-hf

SGLang

How to use state-spaces/mamba-790m-hf 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 "state-spaces/mamba-790m-hf" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "state-spaces/mamba-790m-hf",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

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 "state-spaces/mamba-790m-hf" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "state-spaces/mamba-790m-hf",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use state-spaces/mamba-790m-hf with Docker Model Runner:
```
docker model run hf.co/state-spaces/mamba-790m-hf
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

Mamba

This repository contains the transfromers compatible mamba-2.8b. The checkpoints are untouched, but the full config.json and tokenizer are pushed to this repo.

Usage

You need to install transformers from main until transformers=4.39.0 is released.

pip install git+https://github.com/huggingface/transformers@main

We also recommend you to install both causal_conv_1d and mamba-ssm using:

pip install causal-conv1d>=1.2.0
pip install mamba-ssm

If any of these two is not installed, the "eager" implementation will be used. Otherwise the more optimised cuda kernels will be used.

Generation

You can use the classic generate API:

>>> from transformers import MambaConfig, MambaForCausalLM, AutoTokenizer
>>> import torch

>>> tokenizer = AutoTokenizer.from_pretrained("state-spaces/mamba-790m-hf")
>>> model = MambaForCausalLM.from_pretrained("state-spaces/mamba-790m-hf")
>>> input_ids = tokenizer("Hey how are you doing?", return_tensors="pt")["input_ids"]

>>> out = model.generate(input_ids, max_new_tokens=10)
>>> print(tokenizer.batch_decode(out))
["Hey how are you doing?\n\nI'm good.\n\nHow are"]

PEFT finetuning example

In order to finetune using the peft library, we recommend keeping the model in float32!

from datasets import load_dataset
from trl import SFTTrainer
from peft import LoraConfig
from transformers import AutoTokenizer, AutoModelForCausalLM, TrainingArguments
tokenizer = AutoTokenizer.from_pretrained("state-spaces/mamba-790m-hf")
model = AutoModelForCausalLM.from_pretrained("state-spaces/mamba-790m-hf")
dataset = load_dataset("Abirate/english_quotes", split="train")
training_args = TrainingArguments(
    output_dir="./results",
    num_train_epochs=3,
    per_device_train_batch_size=4,
    logging_dir='./logs',
    logging_steps=10,
    learning_rate=2e-3
)
lora_config =  LoraConfig(
        r=8,
        target_modules=["x_proj", "embeddings", "in_proj", "out_proj"],
        task_type="CAUSAL_LM",
        bias="none"
)
trainer = SFTTrainer(
    model=model,
    tokenizer=tokenizer,
    args=training_args,
    peft_config=lora_config,
    train_dataset=dataset,
    dataset_text_field="quote",
)
trainer.train()