Instructions to use codys12/Hunyuen-preview2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use codys12/Hunyuen-preview2 with Transformers:

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

pipe = pipeline("text-generation", model="codys12/Hunyuen-preview2", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("codys12/Hunyuen-preview2", trust_remote_code=True, dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use codys12/Hunyuen-preview2 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "codys12/Hunyuen-preview2"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "codys12/Hunyuen-preview2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/codys12/Hunyuen-preview2

SGLang

How to use codys12/Hunyuen-preview2 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 "codys12/Hunyuen-preview2" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "codys12/Hunyuen-preview2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

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 "codys12/Hunyuen-preview2" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "codys12/Hunyuen-preview2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use codys12/Hunyuen-preview2 with Docker Model Runner:
```
docker model run hf.co/codys12/Hunyuen-preview2
```

Hunyuen-preview2 / config.json

codys12

Upload HunYuanMoEV1ForCausalLM

8abb784 verified 10 months ago

raw

history blame contribute delete

3.35 kB

	{
	"add_classification_head": false,
	"anyres_pooling_size": 2,
	"anyres_vit_max_image_size": null,
	"anyres_vit_two_views": false,
	"architectures": [
	"HunYuanMoEV1ForCausalLM"
	],
	"attention_bias": false,
	"attention_dropout": 0.1,
	"attention_head_dim": 128,
	"auto_map": {
	"AutoConfig": "configuration_hunyuan.HunYuanConfig",
	"AutoModel": "hunyuan.HunYuanModel",
	"AutoModelForCausalLM": "hunyuan.HunYuanMoEV1ForCausalLM"
	},
	"bos_token_id": 1,
	"cla_share_factor": 2,
	"class_num": 0,
	"dense_list": [
	4096,
	0
	],
	"eod_token_id": 127967,
	"eos_token_id": 127960,
	"group_limited_greedy": false,
	"hidden_act": "silu",
	"hidden_size": 4096,
	"im_end_id": 6,
	"im_newline_id": 12,
	"im_start_id": 5,
	"image_token_id": 9,
	"initializer_range": 0.02,
	"intermediate_size": 3072,
	"kv_lora_rank": null,
	"mask_init_id": 13,
	"max_position_embeddings": 32768,
	"mlp_bias": false,
	"model_type": "hunyuan",
	"moe_drop_tokens": false,
	"moe_intermediate_size": [
	3072,
	3072,
	3072,
	3072,
	3072,
	3072,
	3072,
	3072,
	3072,
	3072,
	3072,
	3072,
	3072,
	3072,
	3072,
	3072,
	3072,
	3072,
	3072,
	3072,
	3072,
	3072,
	3072,
	3072,
	3072,
	3072,
	3072,
	3072,
	3072,
	3072,
	3072,
	3072
	],
	"moe_layer_num_skipped": 0,
	"moe_random_routing_dropped_token": false,
	"moe_topk": [
	8,
	8,
	8,
	8,
	8,
	8,
	8,
	8,
	8,
	8,
	8,
	8,
	8,
	8,
	8,
	8,
	8,
	8,
	8,
	8,
	8,
	8,
	8,
	8,
	8,
	8,
	8,
	8,
	8,
	8,
	8,
	8
	],
	"n_group": null,
	"norm_topk_prob": true,
	"norm_type": "rms",
	"num_attention_heads": 32,
	"num_experts": 64,
	"num_hidden_layers": 32,
	"num_key_value_heads": 8,
	"num_media_embeds": 257,
	"num_shared_expert": [
	1,
	1,
	1,
	1,
	1,
	1,
	1,
	1,
	1,
	1,
	1,
	1,
	1,
	1,
	1,
	1,
	1,
	1,
	1,
	1,
	1,
	1,
	1,
	1,
	1,
	1,
	1,
	1,
	1,
	1,
	1,
	1
	],
	"org_vocab_size": 128167,
	"pad_id": 127961,
	"pad_token_id": 127961,
	"pool_type": "last",
	"position_embedding_xdrope": false,
	"pretraining_tp": 1,
	"q_lora_rank": null,
	"qk_nope_head_dim": null,
	"qk_rope_head_dim": null,
	"rms_norm_eps": 1e-05,
	"rope_scaling": {
	"alpha": 1000.0,
	"beta_fast": 32,
	"beta_slow": 1,
	"factor": 1.0,
	"mscale": 1.0,
	"mscale_all_dim": 1.0,
	"type": "dynamic"
	},
	"rope_theta": 10000.0,
	"routed_scaling_factor": 1.0,
	"sep_token_id": 127962,
	"skip_cls_token": false,
	"text_end_id": 8,
	"text_start_id": 7,
	"tie_word_embeddings": true,
	"topk_group": null,
	"torch_dtype": "float16",
	"transformers_version": "4.52.4",
	"use_cache": true,
	"use_cla": false,
	"use_mixed_mlp_moe": true,
	"use_mla": false,
	"use_qk_norm": true,
	"use_rotary_pos_emb": true,
	"v_head_dim": null,
	"video_end_id": 11,
	"video_start_id": 10,
	"vit_add_patchemb_bias": false,
	"vit_input_resolution": 224,
	"vit_mapping_type": "resampler",
	"vit_norm_type": "fused",
	"vit_patch": 1,
	"vit_path": null,
	"vit_remove_prenorm": false,
	"vit_token": 64,
	"vit_type": null,
	"vit_used_rms_norm": false,
	"vocab_size": 128167,
	"xdrope_section": null
	}