Instructions to use dotan1111/BetaInfer_Configuration3 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use dotan1111/BetaInfer_Configuration3 with Transformers:

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

pipe = pipeline("text-generation", model="dotan1111/BetaInfer_Configuration3")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("dotan1111/BetaInfer_Configuration3")
model = AutoModelForCausalLM.from_pretrained("dotan1111/BetaInfer_Configuration3", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use dotan1111/BetaInfer_Configuration3 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "dotan1111/BetaInfer_Configuration3"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "dotan1111/BetaInfer_Configuration3",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/dotan1111/BetaInfer_Configuration3

SGLang

How to use dotan1111/BetaInfer_Configuration3 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 "dotan1111/BetaInfer_Configuration3" \
    --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": "dotan1111/BetaInfer_Configuration3",
		"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 "dotan1111/BetaInfer_Configuration3" \
        --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": "dotan1111/BetaInfer_Configuration3",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use dotan1111/BetaInfer_Configuration3 with Docker Model Runner:
```
docker model run hf.co/dotan1111/BetaInfer_Configuration3
```

BetaInfer_Configuration3 / config.json

dotan1111

Upload ZambaForCausalLM

d87678b verified about 7 hours ago

Raw

History Blame Contribute Delete

1.37 kB

	{
	"_name_or_path": "/groups/pupko/edodotan/BetaAlignPart3/test_architecture/mix_training_state_space_models_10_15_0p200_phylogenetic_trees/ZAMBA/layers_12_d_1408_e-4_ZAMBA/checkpoint-20480/",
	"architectures": [
	"ZambaForCausalLM"
	],
	"attention_dropout": 0.0,
	"attention_head_dim": 176,
	"attention_hidden_size": 2816,
	"attn_layer_offset": 4,
	"attn_layer_period": 6,
	"bos_token_id": 1,
	"eos_token_id": 4,
	"hidden_act": "gelu",
	"hidden_mamba_act": "silu",
	"hidden_size": 1408,
	"initializer_range": 0.02,
	"intermediate_size": 4928,
	"layers_block_type": [
	"mamba",
	"mamba",
	"hybrid",
	"mamba",
	"mamba",
	"mamba",
	"mamba",
	"hybrid",
	"mamba",
	"mamba",
	"mamba",
	"mamba"
	],
	"mamba_conv_bias": true,
	"mamba_d_conv": 4,
	"mamba_d_state": 16,
	"mamba_dt_rank": 88,
	"mamba_expand": 2,
	"mamba_proj_bias": false,
	"max_position_embeddings": 2048,
	"model_type": "zamba",
	"n_mamba_heads": 2,
	"num_attention_heads": 16,
	"num_heads": 44,
	"num_hidden_layers": 12,
	"num_key_value_heads": 16,
	"num_logits_to_keep": 1,
	"pad_token_id": 3,
	"rms_norm_eps": 1e-05,
	"time_step_floor": 0.0001,
	"time_step_max": 0.1,
	"time_step_min": 0.001,
	"torch_dtype": "float32",
	"transformers_version": "4.48.1",
	"use_cache": true,
	"use_mamba_kernels": true,
	"vocab_size": 6400
	}