Instructions to use DedeProGames/DynamicMind-MoE-Instruct with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use DedeProGames/DynamicMind-MoE-Instruct with Transformers:

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

pipe = pipeline("text-generation", model="DedeProGames/DynamicMind-MoE-Instruct", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("DedeProGames/DynamicMind-MoE-Instruct", trust_remote_code=True, device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use DedeProGames/DynamicMind-MoE-Instruct with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/DedeProGames/DynamicMind-MoE-Instruct

SGLang

How to use DedeProGames/DynamicMind-MoE-Instruct 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 "DedeProGames/DynamicMind-MoE-Instruct" \
    --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": "DedeProGames/DynamicMind-MoE-Instruct",
		"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 "DedeProGames/DynamicMind-MoE-Instruct" \
        --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": "DedeProGames/DynamicMind-MoE-Instruct",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use DedeProGames/DynamicMind-MoE-Instruct with Docker Model Runner:
```
docker model run hf.co/DedeProGames/DynamicMind-MoE-Instruct
```

DynamicMind-MoE-Instruct / config.json

DedeProGames

Add DynamicMind-MoE-Instruct: 2-epoch SFT on smol-smoltalk, custom MoE code, EOS-weighted loss

8d4372c verified 15 days ago

Raw History Blame Contribute Delete

1.03 kB

{
  "architectures": [
    "DynamicMindMoEForCausalLM"
  ],
  "attention_dropout": 0.0,
  "auto_map": {
    "AutoConfig": "configuration_dynamicmind_moe.DynamicMindMoEConfig",
    "AutoModelForCausalLM": "modeling_dynamicmind_moe.DynamicMindMoEForCausalLM"
  },
  "bos_token_id": 0,
  "dtype": "bfloat16",
  "eos_token_id": 0,
  "first_k_dense_layers": 0,
  "hidden_size": 256,
  "intermediate_size": 768,
  "max_position_embeddings": 1024,
  "model_type": "dynamicmind_moe",
  "moe_intermediate_size": 256,
  "norm_topk_prob": true,
  "num_attention_heads": 8,
  "num_experts_per_token": 2,
  "num_hidden_layers": 9,
  "num_key_value_heads": 2,
  "num_routed_experts": 14,
  "num_shared_experts": 1,
  "pad_token_id": 1,
  "rms_norm_eps": 1e-05,
  "rope_theta": 10000.0,
  "router_aux_loss_coef": 0.01,
  "router_bias_update_rate": 0.001,
  "router_z_loss_coef": 0.001,
  "torch_dtype": "bfloat16",
  "transformers_version": "4.53.3",
  "use_aux_loss_free_balancing": true,
  "vocab_size": 8192
}

	{
	"architectures": [
	"DynamicMindMoEForCausalLM"
	],
	"attention_dropout": 0.0,
	"auto_map": {
	"AutoConfig": "configuration_dynamicmind_moe.DynamicMindMoEConfig",
	"AutoModelForCausalLM": "modeling_dynamicmind_moe.DynamicMindMoEForCausalLM"
	},
	"bos_token_id": 0,
	"dtype": "bfloat16",
	"eos_token_id": 0,
	"first_k_dense_layers": 0,
	"hidden_size": 256,
	"intermediate_size": 768,
	"max_position_embeddings": 1024,
	"model_type": "dynamicmind_moe",
	"moe_intermediate_size": 256,
	"norm_topk_prob": true,
	"num_attention_heads": 8,
	"num_experts_per_token": 2,
	"num_hidden_layers": 9,
	"num_key_value_heads": 2,
	"num_routed_experts": 14,
	"num_shared_experts": 1,
	"pad_token_id": 1,
	"rms_norm_eps": 1e-05,
	"rope_theta": 10000.0,
	"router_aux_loss_coef": 0.01,
	"router_bias_update_rate": 0.001,
	"router_z_loss_coef": 0.001,
	"torch_dtype": "bfloat16",
	"transformers_version": "4.53.3",
	"use_aux_loss_free_balancing": true,
	"vocab_size": 8192
	}