Text Generation
Transformers
hyperlm_qwen3
hypergraph
large-language-models
qwen3
projector
hyper-align
hypergraph-as-language
Instructions to use MengqiLei/hyper-align with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use MengqiLei/hyper-align with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="MengqiLei/hyper-align")# Load model directly from transformers import AutoModelForCausalLM model = AutoModelForCausalLM.from_pretrained("MengqiLei/hyper-align", dtype="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps
- vLLM
How to use MengqiLei/hyper-align with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "MengqiLei/hyper-align" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "MengqiLei/hyper-align", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/MengqiLei/hyper-align
- SGLang
How to use MengqiLei/hyper-align 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 "MengqiLei/hyper-align" \ --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": "MengqiLei/hyper-align", "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 "MengqiLei/hyper-align" \ --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": "MengqiLei/hyper-align", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use MengqiLei/hyper-align with Docker Model Runner:
docker model run hf.co/MengqiLei/hyper-align
| { | |
| "architectures": [ | |
| "Qwen3ForCausalLM" | |
| ], | |
| "attention_bias": false, | |
| "attention_dropout": 0.0, | |
| "bos_token_id": 151643, | |
| "consistency_start_step": 0, | |
| "consistency_warmup_steps": 0, | |
| "eos_token_id": 151645, | |
| "freeze_mm_mlp_adapter": false, | |
| "head_dim": 128, | |
| "hidden_act": "silu", | |
| "hidden_size": 4096, | |
| "htp_num_layers": 1, | |
| "htp_semantic_core_dim": 384, | |
| "htp_structure_sidecar_dim": 64, | |
| "hypergraph_semantic_dim": 1024, | |
| "hypergraph_structure_dim": 24, | |
| "initializer_range": 0.02, | |
| "intermediate_size": 12288, | |
| "lambda_ord": 0.01, | |
| "lambda_rel": 0.01, | |
| "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" | |
| ], | |
| "max_position_embeddings": 40960, | |
| "max_window_layers": 36, | |
| "mm_hidden_size": 1048, | |
| "mm_projector_type": "htp", | |
| "mm_use_graph_special_token": false, | |
| "mm_use_graph_start_end": false, | |
| "model_type": "hyperlm_qwen3", | |
| "num_attention_heads": 32, | |
| "num_hidden_layers": 36, | |
| "num_key_value_heads": 8, | |
| "projector_incidence_mode": "sample_real", | |
| "rms_norm_eps": 1e-06, | |
| "rope_scaling": null, | |
| "rope_theta": 1000000, | |
| "sliding_window": null, | |
| "tie_word_embeddings": false, | |
| "torch_dtype": "float32", | |
| "transformers_version": "4.53.0", | |
| "tune_mm_mlp_adapter": true, | |
| "use_cache": true, | |
| "use_mm_proj": true, | |
| "use_sliding_window": false, | |
| "vocab_size": 151936 | |
| } | |