Instructions to use tiiuae/falcon-mamba-7b-4bit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use tiiuae/falcon-mamba-7b-4bit with Transformers:

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

pipe = pipeline("text-generation", model="tiiuae/falcon-mamba-7b-4bit")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("tiiuae/falcon-mamba-7b-4bit")
model = AutoModelForCausalLM.from_pretrained("tiiuae/falcon-mamba-7b-4bit")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use tiiuae/falcon-mamba-7b-4bit with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "tiiuae/falcon-mamba-7b-4bit"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "tiiuae/falcon-mamba-7b-4bit",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/tiiuae/falcon-mamba-7b-4bit

SGLang

How to use tiiuae/falcon-mamba-7b-4bit 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 "tiiuae/falcon-mamba-7b-4bit" \
    --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": "tiiuae/falcon-mamba-7b-4bit",
		"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 "tiiuae/falcon-mamba-7b-4bit" \
        --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": "tiiuae/falcon-mamba-7b-4bit",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use tiiuae/falcon-mamba-7b-4bit with Docker Model Runner:
```
docker model run hf.co/tiiuae/falcon-mamba-7b-4bit
```

Gkunsch commited on Aug 15, 2024

Commit

062ab70

verified ·

1 Parent(s): 9d89ad3

Update config.json

Browse files

Files changed (1) hide show

config.json +2 -2

config.json CHANGED Viewed

@@ -3,7 +3,7 @@
   "architectures": [
     "FalconMambaForCausalLM"
   ],
-  "bos_token_id": 8,
   "conv_kernel": 4,
   "eos_token_id": 11,
   "expand": 16,
@@ -14,7 +14,7 @@
   "layer_norm_epsilon": 1e-05,
   "model_type": "falcon_mamba",
   "num_hidden_layers": 64,
-  "pad_token_id": 0,
   "quantization_config": {
     "_load_in_4bit": true,
     "_load_in_8bit": false,

   "architectures": [
     "FalconMambaForCausalLM"
   ],
+  "bos_token_id": 0,
   "conv_kernel": 4,
   "eos_token_id": 11,
   "expand": 16,
   "layer_norm_epsilon": 1e-05,
   "model_type": "falcon_mamba",
   "num_hidden_layers": 64,
+  "pad_token_id": 11,
   "quantization_config": {
     "_load_in_4bit": true,
     "_load_in_8bit": false,