Instructions to use bedio/MobileLLM-R1-140M-base_stack_48 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use bedio/MobileLLM-R1-140M-base_stack_48 with Transformers:

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

pipe = pipeline("text-generation", model="bedio/MobileLLM-R1-140M-base_stack_48")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("bedio/MobileLLM-R1-140M-base_stack_48")
model = AutoModelForCausalLM.from_pretrained("bedio/MobileLLM-R1-140M-base_stack_48")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use bedio/MobileLLM-R1-140M-base_stack_48 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "bedio/MobileLLM-R1-140M-base_stack_48"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "bedio/MobileLLM-R1-140M-base_stack_48",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/bedio/MobileLLM-R1-140M-base_stack_48

SGLang

How to use bedio/MobileLLM-R1-140M-base_stack_48 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 "bedio/MobileLLM-R1-140M-base_stack_48" \
    --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": "bedio/MobileLLM-R1-140M-base_stack_48",
		"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 "bedio/MobileLLM-R1-140M-base_stack_48" \
        --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": "bedio/MobileLLM-R1-140M-base_stack_48",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use bedio/MobileLLM-R1-140M-base_stack_48 with Docker Model Runner:
```
docker model run hf.co/bedio/MobileLLM-R1-140M-base_stack_48
```

MobileLLM-R1-140M-base_stack_48 / config.json

bedio

Upload Llama4ForCausalLM

70ba857 verified 8 months ago

raw

history blame contribute delete

3.22 kB

	{
	"architectures": [
	"Llama4ForCausalLM"
	],
	"attention_bias": false,
	"attention_chunk_size": 32768,
	"attention_dropout": 0.0,
	"attn_scale": 0.1,
	"attn_temperature_tuning": false,
	"bos_token_id": 128000,
	"eos_token_id": [
	128001,
	128008,
	128009
	],
	"floor_scale": 8192,
	"for_llm_compressor": false,
	"head_dim": 64,
	"hidden_act": "silu",
	"hidden_size": 576,
	"initializer_range": 0.02,
	"interleave_moe_layer_step": 0,
	"intermediate_size": 8192,
	"intermediate_size_mlp": 2048,
	"layer_types": [
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention"
	],
	"max_position_embeddings": 4096,
	"model_type": "llama4_text",
	"moe_layers": [],
	"no_rope_layers": [
	1,
	1,
	1,
	1,
	1,
	1,
	1,
	1,
	1,
	1,
	1,
	1,
	1,
	1,
	1,
	1,
	1,
	1,
	1,
	1,
	1,
	1,
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention"
	],
	"num_attention_heads": 9,
	"num_experts_per_tok": 0,
	"num_hidden_layers": 48,
	"num_key_value_heads": 3,
	"num_local_experts": 0,
	"output_router_logits": false,
	"rms_norm_eps": 1e-05,
	"rope_scaling": null,
	"rope_theta": 1000000.0,
	"router_aux_loss_coef": 0.001,
	"router_jitter_noise": 0.0,
	"tie_word_embeddings": true,
	"torch_dtype": "float32",
	"transformers_version": "4.55.4",
	"use_cache": true,
	"use_qk_norm": true,
	"vocab_size": 128256
	}