Instructions to use Dracones/c4ai-command-r-v01_exl2_7.0bpw-rpcal with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Dracones/c4ai-command-r-v01_exl2_7.0bpw-rpcal with Transformers:

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

pipe = pipeline("text-generation", model="Dracones/c4ai-command-r-v01_exl2_7.0bpw-rpcal")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Dracones/c4ai-command-r-v01_exl2_7.0bpw-rpcal")
model = AutoModelForCausalLM.from_pretrained("Dracones/c4ai-command-r-v01_exl2_7.0bpw-rpcal", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use Dracones/c4ai-command-r-v01_exl2_7.0bpw-rpcal with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Dracones/c4ai-command-r-v01_exl2_7.0bpw-rpcal"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Dracones/c4ai-command-r-v01_exl2_7.0bpw-rpcal",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/Dracones/c4ai-command-r-v01_exl2_7.0bpw-rpcal

SGLang

How to use Dracones/c4ai-command-r-v01_exl2_7.0bpw-rpcal 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 "Dracones/c4ai-command-r-v01_exl2_7.0bpw-rpcal" \
    --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": "Dracones/c4ai-command-r-v01_exl2_7.0bpw-rpcal",
		"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 "Dracones/c4ai-command-r-v01_exl2_7.0bpw-rpcal" \
        --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": "Dracones/c4ai-command-r-v01_exl2_7.0bpw-rpcal",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use Dracones/c4ai-command-r-v01_exl2_7.0bpw-rpcal with Docker Model Runner:
```
docker model run hf.co/Dracones/c4ai-command-r-v01_exl2_7.0bpw-rpcal
```

c4ai-command-r-v01_exl2_7.0bpw-rpcal / config.json

Dracones

Upload folder using huggingface_hub

575ca26 verified over 2 years ago

Raw

History Blame Contribute Delete

1.1 kB

	{
	"_name_or_path": "/home/ahmet_cohere_com/HF_Final_weight_tie",
	"architectures": [
	"CohereForCausalLM"
	],
	"attention_bias": false,
	"attention_dropout": 0.0,
	"bos_token_id": 5,
	"eos_token_id": 255001,
	"hidden_act": "silu",
	"hidden_size": 8192,
	"initializer_range": 0.02,
	"intermediate_size": 22528,
	"layer_norm_eps": 1e-05,
	"logit_scale": 0.0625,
	"max_position_embeddings": 8192,
	"model_max_length": 131072,
	"model_type": "cohere",
	"num_attention_heads": 64,
	"num_hidden_layers": 40,
	"num_key_value_heads": 64,
	"pad_token_id": 0,
	"pretraining_tp": 1,
	"rope_theta": 8000000.0,
	"torch_dtype": "float16",
	"transformers_version": "4.38.2",
	"use_cache": true,
	"vocab_size": 256000,
	"tie_word_embeddings": true,
	"quantization_config": {
	"quant_method": "exl2",
	"version": "0.0.18",
	"bits": 7.0,
	"head_bits": 6,
	"calibration": {
	"rows": 100,
	"length": 2048,
	"dataset": "pippa_raw_fix.parquet"
	}
	}
	}