Instructions to use maldv/Eva-Mindlink-72b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use maldv/Eva-Mindlink-72b with Transformers:

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

pipe = pipeline("text-generation", model="maldv/Eva-Mindlink-72b")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("maldv/Eva-Mindlink-72b")
model = AutoModelForCausalLM.from_pretrained("maldv/Eva-Mindlink-72b")
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 Settings

vLLM

How to use maldv/Eva-Mindlink-72b with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "maldv/Eva-Mindlink-72b"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "maldv/Eva-Mindlink-72b",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/maldv/Eva-Mindlink-72b

SGLang

How to use maldv/Eva-Mindlink-72b 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 "maldv/Eva-Mindlink-72b" \
    --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": "maldv/Eva-Mindlink-72b",
		"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 "maldv/Eva-Mindlink-72b" \
        --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": "maldv/Eva-Mindlink-72b",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use maldv/Eva-Mindlink-72b with Docker Model Runner:
```
docker model run hf.co/maldv/Eva-Mindlink-72b
```

Eva-Mindlink-72b / merge_progress.json

maldv

Upload folder using huggingface_hub

028fcc3 verified 10 months ago

raw

history blame contribute delete

2.64 kB

{"processed_tensors": ["model.layers.76.mlp.up_proj.weight", "model.layers.75.mlp.up_proj.weight", "model.layers.76.mlp.gate_proj.weight", "model.layers.78.input_layernorm.weight", "model.layers.79.self_attn.v_proj.weight", "model.layers.76.self_attn.k_proj.bias", "model.layers.78.self_attn.o_proj.weight", "model.layers.76.self_attn.v_proj.bias", "model.layers.75.self_attn.v_proj.bias", "model.layers.76.self_attn.k_proj.weight", "model.layers.78.self_attn.k_proj.bias", "model.layers.76.self_attn.q_proj.weight", "model.layers.78.self_attn.q_proj.bias", "model.layers.75.self_attn.q_proj.bias", "model.layers.79.self_attn.k_proj.weight", "model.layers.77.mlp.gate_proj.weight", "model.layers.79.self_attn.v_proj.bias", "model.layers.75.post_attention_layernorm.weight", "model.layers.76.self_attn.v_proj.weight", "model.layers.79.self_attn.o_proj.weight", "model.layers.77.input_layernorm.weight", "model.layers.78.mlp.down_proj.weight", "model.layers.79.mlp.down_proj.weight", "model.layers.77.self_attn.k_proj.bias", "model.layers.75.mlp.down_proj.weight", "model.layers.75.mlp.gate_proj.weight", "model.layers.78.mlp.up_proj.weight", "model.layers.77.self_attn.v_proj.bias", "model.layers.78.self_attn.k_proj.weight", "model.layers.77.mlp.down_proj.weight", "model.layers.75.self_attn.o_proj.weight", "model.layers.76.self_attn.q_proj.bias", "model.layers.78.mlp.gate_proj.weight", "model.layers.77.self_attn.o_proj.weight", "model.layers.78.self_attn.v_proj.weight", "model.layers.77.self_attn.v_proj.weight", "model.layers.79.post_attention_layernorm.weight", "model.norm.weight", "model.layers.79.self_attn.q_proj.bias", "model.layers.78.self_attn.v_proj.bias", "model.layers.77.post_attention_layernorm.weight", "model.layers.79.mlp.gate_proj.weight", "model.layers.79.input_layernorm.weight", "model.layers.77.mlp.up_proj.weight", "model.layers.76.mlp.down_proj.weight", "model.layers.76.post_attention_layernorm.weight", "model.layers.77.self_attn.q_proj.bias", "model.layers.79.self_attn.k_proj.bias", "lm_head.weight", "model.layers.75.self_attn.k_proj.weight", "model.layers.76.input_layernorm.weight", "model.layers.79.self_attn.q_proj.weight", "model.layers.77.self_attn.q_proj.weight", "model.layers.75.input_layernorm.weight", "model.layers.78.post_attention_layernorm.weight", "model.layers.77.self_attn.k_proj.weight", "model.layers.75.self_attn.v_proj.weight", "model.layers.79.mlp.up_proj.weight", "model.layers.75.self_attn.k_proj.bias", "model.layers.76.self_attn.o_proj.weight", "model.layers.78.self_attn.q_proj.weight", "model.layers.75.self_attn.q_proj.weight"], "last_updated": "2025-08-05T13:36:03.647376"}