Instructions to use IggoOnCode/mamba-2.8b-slimpj-OpenOrca_1ep with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use IggoOnCode/mamba-2.8b-slimpj-OpenOrca_1ep with Transformers:

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

pipe = pipeline("text-generation", model="IggoOnCode/mamba-2.8b-slimpj-OpenOrca_1ep")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("IggoOnCode/mamba-2.8b-slimpj-OpenOrca_1ep")
model = AutoModelForCausalLM.from_pretrained("IggoOnCode/mamba-2.8b-slimpj-OpenOrca_1ep", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use IggoOnCode/mamba-2.8b-slimpj-OpenOrca_1ep with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "IggoOnCode/mamba-2.8b-slimpj-OpenOrca_1ep"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "IggoOnCode/mamba-2.8b-slimpj-OpenOrca_1ep",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/IggoOnCode/mamba-2.8b-slimpj-OpenOrca_1ep

SGLang

How to use IggoOnCode/mamba-2.8b-slimpj-OpenOrca_1ep 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 "IggoOnCode/mamba-2.8b-slimpj-OpenOrca_1ep" \
    --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": "IggoOnCode/mamba-2.8b-slimpj-OpenOrca_1ep",
		"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 "IggoOnCode/mamba-2.8b-slimpj-OpenOrca_1ep" \
        --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": "IggoOnCode/mamba-2.8b-slimpj-OpenOrca_1ep",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use IggoOnCode/mamba-2.8b-slimpj-OpenOrca_1ep with Docker Model Runner:
```
docker model run hf.co/IggoOnCode/mamba-2.8b-slimpj-OpenOrca_1ep
```

IggoOnCode commited on Mar 8, 2024

Commit

2bf5fee

1 Parent(s): b44e736

transformers compatibility

Browse files

Files changed (1) hide show

config.json +18 -1

config.json CHANGED Viewed

	@@ -1 +1,18 @@
1	- {~~"d_model": 2560, "n_layer": 64, "vocab_size": 50277, "ssm_cfg": {}, "rms_norm": true, "residual_in_fp32": true, "fused_add_norm": true, "pad_vocab_size_multiple": 8}~~

+{
+  "_name_or_path": "IggoOnCode/mamba-2.8b-slimpj-OpenOrca_1ep",
+  "architectures": [
+    "MambaForCausalLM"
+  ],
+  "d_inner": 1536,
+  "d_model": 2560,
+  "fused_add_norm": true,
+  "hidden_size": 2560,
+  "model_type": "mamba",
+  "n_layer": 64,
+  "pad_vocab_size_multiple": 8,
+  "residual_in_fp32": true,
+  "rms_norm": true,
+  "ssm_cfg": {},
+  "torch_dtype": "bfloat16",
+  "vocab_size": 50280
+}