Instructions to use Undi95/MiquTotalMix-70B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Undi95/MiquTotalMix-70B with Transformers:

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

pipe = pipeline("text-generation", model="Undi95/MiquTotalMix-70B")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Undi95/MiquTotalMix-70B")
model = AutoModelForCausalLM.from_pretrained("Undi95/MiquTotalMix-70B")
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 Undi95/MiquTotalMix-70B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Undi95/MiquTotalMix-70B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Undi95/MiquTotalMix-70B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Undi95/MiquTotalMix-70B

SGLang

How to use Undi95/MiquTotalMix-70B 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 "Undi95/MiquTotalMix-70B" \
    --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": "Undi95/MiquTotalMix-70B",
		"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 "Undi95/MiquTotalMix-70B" \
        --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": "Undi95/MiquTotalMix-70B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Undi95/MiquTotalMix-70B with Docker Model Runner:
```
docker model run hf.co/Undi95/MiquTotalMix-70B
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

MiquTotalMix-70B

This is a merge of pre-trained language models created using mergekit.

Merge Details

Merge Method

This model was merged using the passthrough merge method.

Models Merged

The following models were included in the merge:

Prompt template

Prompt template 1: Alpaca (try this first)

### Input:
{system prompt}

### Instruction:
{prompt}

### Response:

Prompt template 2: Mistral

[INST] {prompt} [/INST]

Configuration

The following YAML configuration was used to produce this model:

merge_method: passthrough
slices:
- sources:
  - layer_range: [0, 1]
    model:
      model:
        path: sophosympatheia/Midnight-Miqu-70B-v1.0
- sources:
  - layer_range: [1, 2]
    model:
      model:
        path: NeverSleep/MiquMaid-v1-70B
- sources:
  - layer_range: [2, 3]
    model:
      model:
        path: 152334H/miqu-1-70b-sf
- sources:
  - layer_range: [3, 4]
    model:
      model:
        path: 152334H/miqu-1-70b-sf
- sources:
  - layer_range: [4, 5]
    model:
      model:
        path: NeverSleep/MiquMaid-v1-70B
- sources:
  - layer_range: [5, 6]
    model:
      model:
        path: 152334H/miqu-1-70b-sf
- sources:
  - layer_range: [6, 7]
    model:
      model:
        path: sophosympatheia/Midnight-Miqu-70B-v1.0
- sources:
  - layer_range: [7, 8]
    model:
      model:
        path: sophosympatheia/Midnight-Miqu-70B-v1.0
- sources:
  - layer_range: [8, 9]
    model:
      model:
        path: 152334H/miqu-1-70b-sf
- sources:
  - layer_range: [9, 10]
    model:
      model:
        path: sophosympatheia/Midnight-Miqu-70B-v1.0
- sources:
  - layer_range: [10, 11]
    model:
      model:
        path: alchemonaut/BoreanGale-70B
- sources:
  - layer_range: [11, 12]
    model:
      model:
        path: sophosympatheia/Midnight-Miqu-70B-v1.0
- sources:
  - layer_range: [12, 13]
    model:
      model:
        path: 152334H/miqu-1-70b-sf
- sources:
  - layer_range: [13, 14]
    model:
      model:
        path: alchemonaut/BoreanGale-70B
- sources:
  - layer_range: [14, 15]
    model:
      model:
        path: 152334H/miqu-1-70b-sf
- sources:
  - layer_range: [15, 16]
    model:
      model:
        path: alchemonaut/BoreanGale-70B
- sources:
  - layer_range: [16, 17]
    model:
      model:
        path: NeverSleep/MiquMaid-v1-70B
- sources:
  - layer_range: [17, 18]
    model:
      model:
        path: 152334H/miqu-1-70b-sf
- sources:
  - layer_range: [18, 19]
    model:
      model:
        path: sophosympatheia/Midnight-Miqu-70B-v1.0
- sources:
  - layer_range: [19, 20]
    model:
      model:
        path: 152334H/miqu-1-70b-sf
- sources:
  - layer_range: [20, 21]
    model:
      model:
        path: alchemonaut/BoreanGale-70B
- sources:
  - layer_range: [21, 22]
    model:
      model:
        path: 152334H/miqu-1-70b-sf
- sources:
  - layer_range: [22, 23]
    model:
      model:
        path: 152334H/miqu-1-70b-sf
- sources:
  - layer_range: [23, 24]
    model:
      model:
        path: sophosympatheia/Midnight-Miqu-70B-v1.0
- sources:
  - layer_range: [24, 25]
    model:
      model:
        path: alchemonaut/BoreanGale-70B
- sources:
  - layer_range: [25, 26]
    model:
      model:
        path: NeverSleep/MiquMaid-v1-70B
- sources:
  - layer_range: [26, 27]
    model:
      model:
        path: sophosympatheia/Midnight-Miqu-70B-v1.0
- sources:
  - layer_range: [27, 28]
    model:
      model:
        path: 152334H/miqu-1-70b-sf
- sources:
  - layer_range: [28, 29]
    model:
      model:
        path: 152334H/miqu-1-70b-sf
- sources:
  - layer_range: [29, 30]
    model:
      model:
        path: NeverSleep/MiquMaid-v1-70B
- sources:
  - layer_range: [30, 31]
    model:
      model:
        path: NeverSleep/MiquMaid-v1-70B
- sources:
  - layer_range: [31, 32]
    model:
      model:
        path: NeverSleep/MiquMaid-v1-70B
- sources:
  - layer_range: [32, 33]
    model:
      model:
        path: alchemonaut/BoreanGale-70B
- sources:
  - layer_range: [33, 34]
    model:
      model:
        path: alchemonaut/BoreanGale-70B
- sources:
  - layer_range: [34, 35]
    model:
      model:
        path: alchemonaut/BoreanGale-70B
- sources:
  - layer_range: [35, 36]
    model:
      model:
        path: sophosympatheia/Midnight-Miqu-70B-v1.0
- sources:
  - layer_range: [36, 37]
    model:
      model:
        path: alchemonaut/BoreanGale-70B
- sources:
  - layer_range: [37, 38]
    model:
      model:
        path: sophosympatheia/Midnight-Miqu-70B-v1.0
- sources:
  - layer_range: [38, 39]
    model:
      model:
        path: 152334H/miqu-1-70b-sf
- sources:
  - layer_range: [39, 40]
    model:
      model:
        path: 152334H/miqu-1-70b-sf
- sources:
  - layer_range: [40, 41]
    model:
      model:
        path: NeverSleep/MiquMaid-v1-70B
- sources:
  - layer_range: [41, 42]
    model:
      model:
        path: 152334H/miqu-1-70b-sf
- sources:
  - layer_range: [42, 43]
    model:
      model:
        path: alchemonaut/BoreanGale-70B
- sources:
  - layer_range: [43, 44]
    model:
      model:
        path: NeverSleep/MiquMaid-v1-70B
- sources:
  - layer_range: [44, 45]
    model:
      model:
        path: sophosympatheia/Midnight-Miqu-70B-v1.0
- sources:
  - layer_range: [45, 46]
    model:
      model:
        path: 152334H/miqu-1-70b-sf
- sources:
  - layer_range: [46, 47]
    model:
      model:
        path: alchemonaut/BoreanGale-70B
- sources:
  - layer_range: [47, 48]
    model:
      model:
        path: 152334H/miqu-1-70b-sf
- sources:
  - layer_range: [48, 49]
    model:
      model:
        path: sophosympatheia/Midnight-Miqu-70B-v1.0
- sources:
  - layer_range: [49, 50]
    model:
      model:
        path: alchemonaut/BoreanGale-70B
- sources:
  - layer_range: [50, 51]
    model:
      model:
        path: alchemonaut/BoreanGale-70B
- sources:
  - layer_range: [51, 52]
    model:
      model:
        path: sophosympatheia/Midnight-Miqu-70B-v1.0
- sources:
  - layer_range: [52, 53]
    model:
      model:
        path: 152334H/miqu-1-70b-sf
- sources:
  - layer_range: [53, 54]
    model:
      model:
        path: sophosympatheia/Midnight-Miqu-70B-v1.0
- sources:
  - layer_range: [54, 55]
    model:
      model:
        path: 152334H/miqu-1-70b-sf
- sources:
  - layer_range: [55, 56]
    model:
      model:
        path: alchemonaut/BoreanGale-70B
- sources:
  - layer_range: [56, 57]
    model:
      model:
        path: sophosympatheia/Midnight-Miqu-70B-v1.0
- sources:
  - layer_range: [57, 58]
    model:
      model:
        path: NeverSleep/MiquMaid-v1-70B
- sources:
  - layer_range: [58, 59]
    model:
      model:
        path: alchemonaut/BoreanGale-70B
- sources:
  - layer_range: [59, 60]
    model:
      model:
        path: alchemonaut/BoreanGale-70B
- sources:
  - layer_range: [60, 61]
    model:
      model:
        path: sophosympatheia/Midnight-Miqu-70B-v1.0
- sources:
  - layer_range: [61, 62]
    model:
      model:
        path: alchemonaut/BoreanGale-70B
- sources:
  - layer_range: [62, 63]
    model:
      model:
        path: NeverSleep/MiquMaid-v1-70B
- sources:
  - layer_range: [63, 64]
    model:
      model:
        path: 152334H/miqu-1-70b-sf
- sources:
  - layer_range: [64, 65]
    model:
      model:
        path: 152334H/miqu-1-70b-sf
- sources:
  - layer_range: [65, 66]
    model:
      model:
        path: NeverSleep/MiquMaid-v1-70B
- sources:
  - layer_range: [66, 67]
    model:
      model:
        path: sophosympatheia/Midnight-Miqu-70B-v1.0
- sources:
  - layer_range: [67, 68]
    model:
      model:
        path: alchemonaut/BoreanGale-70B
- sources:
  - layer_range: [68, 69]
    model:
      model:
        path: sophosympatheia/Midnight-Miqu-70B-v1.0
- sources:
  - layer_range: [69, 70]
    model:
      model:
        path: 152334H/miqu-1-70b-sf
- sources:
  - layer_range: [70, 71]
    model:
      model:
        path: alchemonaut/BoreanGale-70B
- sources:
  - layer_range: [71, 72]
    model:
      model:
        path: alchemonaut/BoreanGale-70B
- sources:
  - layer_range: [72, 73]
    model:
      model:
        path: sophosympatheia/Midnight-Miqu-70B-v1.0
- sources:
  - layer_range: [73, 74]
    model:
      model:
        path: NeverSleep/MiquMaid-v1-70B
- sources:
  - layer_range: [74, 75]
    model:
      model:
        path: NeverSleep/MiquMaid-v1-70B
- sources:
  - layer_range: [75, 76]
    model:
      model:
        path: alchemonaut/BoreanGale-70B
- sources:
  - layer_range: [76, 77]
    model:
      model:
        path: 152334H/miqu-1-70b-sf
- sources:
  - layer_range: [77, 78]
    model:
      model:
        path: alchemonaut/BoreanGale-70B
- sources:
  - layer_range: [78, 79]
    model:
      model:
        path: alchemonaut/BoreanGale-70B
- sources:
  - layer_range: [79, 80]
    model:
      model:
        path: 152334H/miqu-1-70b-sf

Downloads last month: 5

Safetensors

Model size

69B params

Tensor type

F32

BF16

F16

Model tree for Undi95/MiquTotalMix-70B

152334H/miqu-1-70b-sf

NeverSleep/MiquMaid-v1-70B

alchemonaut/BoreanGale-70B

sophosympatheia/Midnight-Miqu-70B-v1.0

Merge model

this model

Quantizations

2 models