Nous Hermes 2 Mixtral 8x7B DPO 5.0bpw H6 EXL2 Rpcal is an open-source language model by intervitens. Features: LLM, VRAM: 29.5GB, Context: 32K, License: apache-2.0, MoE, Quantized, HF Score: 65.8, LLM Explorer Score: 0.13, Arc: 92.1, HellaSwag: 81.9, MMLU: 71.8, TruthfulQA: 47.6, WinoGrande: 80.2, GSM8K: 66.
Base model:finetune:mistralai/... Base model:mistralai/mixtral-8... Chatml Conversational Distillation Dpo En Endpoints compatible Exl2 Finetuned Gpt4 Instruct Mixtral Moe Quantized Region:us Rlhf Sharded Synthetic data Tensorflow
Model Card on HF 🤗: https://huggingface.co/intervitens-archive/Nous-Hermes-2-Mixtral-8x7B-DPO-5.0bpw-h6-exl2-rpcal
Nous Hermes 2 Mixtral 8x7B DPO 5.0bpw H6 EXL2 Rpcal Benchmarks
ARC: 92.1 vs 96.7 (so35)-4.8%
HellaSwag: 81.9 vs 95.3 (gpt4)-14.1%
MMLU: 71.8 vs 88.3 (so35)-18.7%
TruthfulQA: 47.62 vs 59 (gpt4)-19.3%
WinoGrande: 80.19 vs 87.5 (gpt4)-8.4%
GSM8K: 66 vs 96.4 (so35)-31.5%
OpenCompas: 59.6 vs 73.3 (gpt4)-18.7%
Exam: 72.7 vs 77.2 (gpt4)-5.8%
Language: 45.2 vs 62 (gpt4)-27.1%
Knowledge: 63.2 vs 73.5 (gpt4)-14%
Understand: 68 vs 70 (gpt4)-2.9%
Reasoning: 55.9 vs 74.4 (gpt4)-24.9%
LLME Score: 0.12759
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)
Nous Hermes 2 Mixtral 8x7B DPO 5.0bpw H6 EXL2 Rpcal Parameters and Internals
| LLM Name | Nous Hermes 2 Mixtral 8x7B DPO 5.0bpw H6 EXL2 Rpcal |
| Repository 🤗 | https://huggingface.co/intervitens-archive/Nous-Hermes-2-Mixtral-8x7B-DPO-5.0bpw-h6-exl2-rpcal |
| Base Model(s) | |
| Required VRAM | 29.5 GB |
| Updated | 2026-04-26 |
| Maintainer | intervitens |
| Model Type | mixtral |
| Model Files | |
| Supported Languages | en |
| Quantization Type | exl2 |
| Model Architecture | MixtralForCausalLM |
| License | apache-2.0 |
| Context Length | 32768 |
| Model Max Length | 32768 |
| Transformers Version | 4.37.0.dev0 |
| Tokenizer Class | LlamaTokenizer |
| Padding Token | </s> |
| Vocabulary Size | 32002 |
| Torch Data Type | bfloat16 |
Best Alternatives to Nous Hermes 2 Mixtral 8x7B DPO 5.0bpw H6 EXL2 Rpcal
Best Alternatives |
Context / RAM |
Downloads |
Likes |
|---|---|---|---|
| ...oE V0.1 DPO F16 4.0bpw H6 EXL2 | 195K / 31.3 GB | 7 | 0 |
| ...oE V0.1 DPO F16 5.0bpw H6 EXL2 | 195K / 38.8 GB | 6 | 0 |
| ...2 Mixtral 8x22b 6.0bpw H8 EXL2 | 64K / 105.8 GB | 2 | 1 |
| WizardLM 2 8x22 EXL2 4.0bpw | 64K / 70.9 GB | 7 | 1 |
| ...M 2 8x22B Beige 5.0bpw H6 EXL2 | 64K / 88.5 GB | 11 | 0 |
| ...M 2 8x22B Beige 2.4bpw H6 EXL2 | 64K / 42.7 GB | 6 | 0 |
| ...M 2 8x22B Beige 3.0bpw H6 EXL2 | 64K / 53.2 GB | 6 | 0 |
| ...rdLM 2 8x22B Beige EXL2 5.0bpw | 64K / 88.4 GB | 6 | 0 |
| ...M 2 8x22B Beige 4.0bpw H6 EXL2 | 64K / 70.8 GB | 5 | 0 |
| ...B Instruct V0.1 8.0bpw H8 EXL2 | 64K / 120.2 GB | 4 | 1 |
Note: green Score (e.g. "73.2") means that the model is better than intervitens-archive/Nous-Hermes-2-Mixtral-8x7B-DPO-5.0bpw-h6-exl2-rpcal .
Rank the Nous Hermes 2 Mixtral 8x7B DPO 5.0bpw H6 EXL2 Rpcal 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? 53254 in total.
Original data from HuggingFace, OpenCompass and various public git repos.
Check out Ag3ntum — our secure, self-hosted AI agent for server management.
Release v20260328a