---
title: MIST-1-70B-MLX
canonical_url: "https://www.modelscope.cn/models/mlx-community/MIST-1-70B-MLX"
md_url: "https://www.modelscope.cn/models/mlx-community/MIST-1-70B-MLX.md"
repository: mlx-community/MIST-1-70B-MLX
last_updated: 2026-05-31
license: llama3.1
pipeline_tag: text-generation
tasks:
  - text-generation
model_type:
  - llama
architectures:
  - LlamaForCausalLM
base_model:
  - nvidia/Llama-3.1-Nemotron-70B-Instruct-HF
  - deepseek-ai/DeepSeek-R1-Distill-Llama-70B
  - NousResearch/Hermes-3-Llama-3.1-70B
  - meta-llama/Llama-3.1-70B-Instruct
base_model_relation: merge
parameters: 70.6B
tensor_type:
  - BF16
library_name:
  - pytorch
  - transformer
  - safetensors
frameworks:
  - pytorch
language:
  - en
downloads: 7
stars: 0
tags:
  - merge
  - dare_ties
  - llama
  - llama-3.1
  - mist
---

# MIST-1-70B-MLX

> MIST-1-70B-MLX - mlx-community 在 ModelScope 开源的模型。MIST-1-70B is the mid-size model in the MIST model family by olaverse. Built by blending 4 of the best Llama 3.1 70B models using DARE+TIES. structured, detailed, production ready

mlx-community/MIST-1-70B-MLX 是 ModelScope 魔搭社区上的 70.6B 参数text-generation模型，采用 llama3.1 许可，基于 nvidia/Llama-3.1-Nemotron-70B-Instruct-HF、deepseek-ai/DeepSeek-R1-Distill-Llama-70B、NousResearch/Hermes-3-Llama-3.1-70B 构建。

- **Repository**: mlx-community/MIST-1-70B-MLX
- **License**: llama3.1
- **Tasks**: text-generation
- **Parameters**: 70.6B
- **Base model**: nvidia/Llama-3.1-Nemotron-70B-Instruct-HF, deepseek-ai/DeepSeek-R1-Distill-Llama-70B, NousResearch/Hermes-3-Llama-3.1-70B, meta-llama/Llama-3.1-70B-Instruct
- **Tags**: merge, dare_ties, llama, llama-3.1, mist
- **Downloads**: 7
- **Stars**: 0
- **Last updated**: 2026-05-31

Source: https://www.modelscope.cn/models/mlx-community/MIST-1-70B-MLX

---

# MIST-1-70B

MIST-1-70B is the mid-size model in the **MIST model family** by [olaverse](https://huggingface.co/olaverse). Built by blending 4 of the best Llama 3.1 70B models using DARE+TIES.
structured, detailed, production ready

## MIST Model Family

| Model | Params | Speed | Status |
|---|---|---|---|
| [MIST-1-8B](https://huggingface.co/olaverse/MIST-Mini-8B) | 8B | ~63 tok/s | ✅ Available |
| **MIST-1-70B** | 70B | ~23 tok/s | ✅ Available |
| [MIST-1-140B](https://huggingface.co/olaverse/MIST-1-140B) | 140B | ~8 tok/s | ✅ Available |

## Key Strengths

- 🧠 **Strong Reasoning** — DeepSeek R1 distillation at 70B scale
- 🤝 **Highly Helpful** — built on Nemotron #1 on helpfulness benchmarks
- 💻 **Coding** — clean documented production-ready code
- 📐 **Math** — step-by-step structured problem solving
- 🌍 **Multilingual** — supports 8+ languages
- 📚 **Long Context** — 128K token context window
- 🔓 **Unrestricted** — follows instructions without excessive refusals

## Merge Method

MIST-1-70B uses **DARE+TIES**:
- **DARE** prunes redundant weights from each model
- **TIES** resolves weight conflicts using sign consensus
- Result: best capabilities of all 4 models combined

## Benchmark Results

| Task | Speed | Quality |
|---|---|---|
| Reasoning | 10.5s | ✅ Correct step-by-step |
| Coding | 11.3s | ✅ Clean with type hints |
| Math | 11.3s | ✅ Structured with verification |
| General | 11.3s | ✅ Accurate and detailed |
| Instruction | 8.1s | ✅ Precise and formatted |

**Average: 23 tok/s**

## How to Use

### bfloat16 — Full Precision (140GB VRAM)
```python
from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained(
    "olaverse/MIST-1-70B",
    torch_dtype="auto",
    device_map="auto",
)
tokenizer = AutoTokenizer.from_pretrained("olaverse/MIST-1-70B")

messages = [{"role": "user", "content": "Your question here"}]
text = tokenizer.apply_chat_template(
    messages, tokenize=False, add_generation_prompt=True
)
inputs = tokenizer(text, return_tensors="pt").to("cuda")
outputs = model.generate(**inputs, max_new_tokens=512)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))
```

### 4-bit Quantized (40GB VRAM)
```python
from transformers import AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfig
import torch

quantization_config = BitsAndBytesConfig(
    load_in_4bit=True,
    bnb_4bit_compute_dtype=torch.bfloat16,
    bnb_4bit_quant_type='nf4'
)
model = AutoModelForCausalLM.from_pretrained(
    "olaverse/MIST-1-70B",
    quantization_config=quantization_config,
    device_map="auto",
)
tokenizer = AutoTokenizer.from_pretrained("olaverse/MIST-1-70B")
```

## Hardware Requirements

| Precision | VRAM | Size |
|---|---|---|
| bfloat16 | 140GB (1x H200 or 2x H100) | 132GB |
| 4-bit (NF4) | 40GB (1x A100/H100) | ~35GB |

## License

[Llama 3.1 Community License](https://llama.meta.com/llama3/license/)