Merged Model Autotrain compatible Conversational Endpoints compatible Mixtral Moe Not-for-all-audiences Nsfw Region:us Safetensors Sharded Tensorflow
Model Card on HF 🤗: https://huggingface.co/NeverSleep/MiquMaid-v2-2x70B-DPO
MiquMaid V2 2x70B DPO Benchmarks
ARC: 72.53 vs 96.7 (so35)-25%
HellaSwag: 88.36 vs 95.3 (gpt4)-7.3%
MMLU: 75.31 vs 88.3 (so35)-14.7%
TruthfulQA: 66.5 vs 59 (gpt4)12.7%
WinoGrande: 85.32 vs 87.5 (gpt4)-2.5%
GSM8K: 67.32 vs 96.4 (so35)-30.2%
LLME Score: 0.14556
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
What is the LLM Explorer Rank (Score)
MiquMaid V2 2x70B DPO Parameters and Internals
| LLM Name | MiquMaid V2 2x70B DPO |
| Repository 🤗 | https://huggingface.co/NeverSleep/MiquMaid-v2-2x70B-DPO |
| Merged Model | Yes |
| Model Size | 125.3b |
| Required VRAM | 251.4 GB |
| Updated | 2025-08-22 |
| Maintainer | NeverSleep |
| Model Type | mixtral |
| Model Files | |
| Model Architecture | MixtralForCausalLM |
| License | cc-by-nc-4.0 |
| Context Length | 32764 |
| Model Max Length | 32764 |
| Transformers Version | 4.37.2 |
| Tokenizer Class | LlamaTokenizer |
| Padding Token | <s> |
| Vocabulary Size | 32000 |
| Torch Data Type | bfloat16 |
Best Alternatives to MiquMaid V2 2x70B DPO
Best Alternatives |
Context / RAM |
Downloads |
Likes |
|---|---|---|---|
| Grafted Wind Elementals 2x70B | 32K / 251.4 GB | 5 | 0 |
| Grafted Llama2 2x70B | 4K / 251.4 GB | 1716 | 3 |
Note: green Score (e.g. "73.2") means that the model is better than NeverSleep/MiquMaid-v2-2x70B-DPO .
Rank the MiquMaid V2 2x70B DPO Capabilities
🆘 Have you tried this model? Rate its performance. This feedback would greatly assist ML community in identifying the most suitable model for their needs. Your contribution really does make a difference! 🌟
What open-source LLMs or SLMs are you in search of? 50835 in total.
Original data from HuggingFace, OpenCompass and various public git repos.
Release v20241124