Instructions to use lightonai/alfred-40b-1023 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use lightonai/alfred-40b-1023 with Transformers:

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

pipe = pipeline("text-generation", model="lightonai/alfred-40b-1023", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("lightonai/alfred-40b-1023", trust_remote_code=True, dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use lightonai/alfred-40b-1023 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "lightonai/alfred-40b-1023"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "lightonai/alfred-40b-1023",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/lightonai/alfred-40b-1023

SGLang

How to use lightonai/alfred-40b-1023 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 "lightonai/alfred-40b-1023" \
    --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": "lightonai/alfred-40b-1023",
		"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 "lightonai/alfred-40b-1023" \
        --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": "lightonai/alfred-40b-1023",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use lightonai/alfred-40b-1023 with Docker Model Runner:
```
docker model run hf.co/lightonai/alfred-40b-1023
```

alfred-40b-1023 / config.json

cthiriet

Add rope_scaling to config.json

cff69f1 verified over 2 years ago

Raw

History Blame Contribute Delete

909 Bytes

	{
	"alibi": false,
	"apply_residual_connection_post_layernorm": false,
	"architectures": [
	"RWForCausalLM"
	],
	"attention_dropout": 0.0,
	"auto_map": {
	"AutoConfig": "configuration_RW.RWConfig",
	"AutoModelForCausalLM": "modeling_RW.RWForCausalLM"
	},
	"bias": false,
	"bos_token_id": 11,
	"embedding_scaling_factor": 4,
	"eos_token_id": 11,
	"hidden_dropout": 0.0,
	"hidden_size": 8192,
	"initializer_range": 0.02,
	"layer_norm_epsilon": 1e-05,
	"model_type": "RefinedWeb",
	"multi_query": true,
	"n_head": 128,
	"n_head_kv": 8,
	"n_layer": 60,
	"ntk_scaling_factor": 5,
	"parallel_attn": true,
	"single_ln": false,
	"torch_dtype": "bfloat16",
	"transformers_version": "4.31.0",
	"use_cache": true,
	"vanilla_scaling_factor": null,
	"vocab_size": 65024,
	"rope_scaling": {
	"type": "ntk_yarn",
	"factor": 4.0,
	"original_max_position_embeddings": 2048
	}
	}