Instructions to use K-intelligence/Midm-2.0-Base-Instruct with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use K-intelligence/Midm-2.0-Base-Instruct with Transformers:

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

pipe = pipeline("text-generation", model="K-intelligence/Midm-2.0-Base-Instruct")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("K-intelligence/Midm-2.0-Base-Instruct")
model = AutoModelForCausalLM.from_pretrained("K-intelligence/Midm-2.0-Base-Instruct")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use K-intelligence/Midm-2.0-Base-Instruct with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "K-intelligence/Midm-2.0-Base-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": "K-intelligence/Midm-2.0-Base-Instruct",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/K-intelligence/Midm-2.0-Base-Instruct

SGLang

How to use K-intelligence/Midm-2.0-Base-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 "K-intelligence/Midm-2.0-Base-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": "K-intelligence/Midm-2.0-Base-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 "K-intelligence/Midm-2.0-Base-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": "K-intelligence/Midm-2.0-Base-Instruct",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use K-intelligence/Midm-2.0-Base-Instruct with Docker Model Runner:
```
docker model run hf.co/K-intelligence/Midm-2.0-Base-Instruct
```

Why Does o_proj Project to hidden_size When num_heads × head_dim ≠ hidden_size?

by ccosmos - opened Sep 9, 2025

Discussion

ccosmos

Sep 9, 2025

Typically, hidden_size is calculated as num_attention_heads × head_dim, but this model’s configuration is as follows:

num_attention_heads: 32

head_dim: 128

hidden_size: 1792

Why does it change the hidden size by projecting to hidden_size in o_proj?

glasshong

Sep 9, 2025

Oh, I also curious about it.

HJ-KT

Sep 11, 2025

Are you referring to the Mi:dm 2.0 Mini model?

The Mi:dm 2.0 Mini is trained through multiple rounds of knowledge distillation and pruning, starting from the Mi:dm-Base (11.5B) model.
To explore ways for the lighter model without sacrificing performance, we ran several experiments with a mix of fixed parameters and variable ones.
As a result, the total number of parameters, number of layers, hidden size, and other parameters were determined.

In short, this architecture was chosen as the outcome of experiments aimed at minimizing performance degradation while reducing model size.
Given that it’s a lightweight model, we prioritized maintaining strong performance over inference-time optimizations.
Thank you.

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment