Instructions to use DeepXR/Helion-V2.5-Rnd with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use DeepXR/Helion-V2.5-Rnd with Transformers:

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

pipe = pipeline("text-generation", model="DeepXR/Helion-V2.5-Rnd", trust_remote_code=True)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("DeepXR/Helion-V2.5-Rnd", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("DeepXR/Helion-V2.5-Rnd", trust_remote_code=True)

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use DeepXR/Helion-V2.5-Rnd with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "DeepXR/Helion-V2.5-Rnd"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "DeepXR/Helion-V2.5-Rnd",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/DeepXR/Helion-V2.5-Rnd

SGLang

How to use DeepXR/Helion-V2.5-Rnd 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 "DeepXR/Helion-V2.5-Rnd" \
    --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": "DeepXR/Helion-V2.5-Rnd",
		"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 "DeepXR/Helion-V2.5-Rnd" \
        --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": "DeepXR/Helion-V2.5-Rnd",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use DeepXR/Helion-V2.5-Rnd with Docker Model Runner:
```
docker model run hf.co/DeepXR/Helion-V2.5-Rnd
```

Trouter-Library Alex Gall commited on Dec 6, 2025

Commit

8878b12

verified ·

1 Parent(s): ed3e1e1

Update model_config.yaml (#6)

Browse files

- Update model_config.yaml (8551ec1adc52c649623c4c23f0388237b5e84a5e)

Co-authored-by: Alex Gall <AlexGall@users.noreply.huggingface.co>

Files changed (1) hide show

model_config.yaml +2 -16

model_config.yaml CHANGED Viewed

@@ -33,7 +33,6 @@ model:
     attention_bias: false
     attention_dropout: 0.0
     mlp_bias: false
   tokenizer:
     type: "sentencepiece"
     model_max_length: 131072
@@ -42,15 +41,6 @@ model:
     chat_template: "{% for message in messages %}{{ '<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>\n' }}{% endfor %}{{ '<|im_start|>assistant\n' }}"
   training:
-    base_model: "meta-llama/Meta-Llama-3.1-70B"
-    training_data:
-      - "scientific_papers"
-      - "code_repositories"
-      - "mathematical_proofs"
-      - "conversational_data"
-      - "multilingual_corpus"
-      - "technical_documentation"
-    total_tokens: "2.5T"
     training_steps: 150000
     warmup_steps: 2000
     learning_rate: 2.0e-5
@@ -70,14 +60,10 @@ model:
   quantization:
     bits: 16
     supported_formats:
       - "fp16"
-      - "bf16"
-      - "int8"
-      - "int4"
-      - "awq"
-      - "gptq"
-      - "gguf"
 inference:
   default_parameters:

     attention_bias: false
     attention_dropout: 0.0
     mlp_bias: false
   tokenizer:
     type: "sentencepiece"
     model_max_length: 131072
     chat_template: "{% for message in messages %}{{ '<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>\n' }}{% endfor %}{{ '<|im_start|>assistant\n' }}"
   training:
     training_steps: 150000
     warmup_steps: 2000
     learning_rate: 2.0e-5
   quantization:
     bits: 16
+    precision: "float16"
     supported_formats:
       - "fp16"
+    note: "Model is provided in full FP16 precision without quantization"
 inference:
   default_parameters: