LLM EXPLORER
 LLM Explorer: A Curated Large Language Model Directory and Analytics
Dark Theme
  Agents   Benchmarks   LLM  List    LLM Hosting    LLM Leaderboards    Blog    Newsfeed    Advertise

Dolphin 2.7 Mixtral 8x7b 8bpw EXL2 by Kearm

 »  All LLMs  »  Kearm  »  Dolphin 2.7 Mixtral 8x7b 8bpw EXL2   URL Share it on

  3   

Dolphin 2.7 Mixtral 8x7b 8bpw EXL2 is an open-source language model by Kearm. Features: LLM, VRAM: 46.8GB, Context: 32K, License: apache-2.0, MoE, Quantized, Instruction-Based, LLM Explorer Score: 0.12.

  Conversational Dataset:cognitivecomputations/... Dataset:cognitivecomputations/... Dataset:ise-uiuc/magicoder-evo... Dataset:ise-uiuc/magicoder-oss... Dataset:jondurbin/airoboros-2....   Dataset:ldjnr/capybara   Dataset:teknium/openhermes   En   Endpoints compatible   Exl2   Instruct   Mixtral   Moe   Pytorch   Quantized   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/Kearm/dolphin-2.7-mixtral-8x7b-8bpw-exl2 

Dolphin 2.7 Mixtral 8x7b 8bpw EXL2 Benchmarks

LLME Score: 0.12265
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Dolphin 2.7 Mixtral 8x7b 8bpw EXL2 (Kearm/dolphin-2.7-mixtral-8x7b-8bpw-exl2)
🌟 Advertise your project 🚀

Dolphin 2.7 Mixtral 8x7b 8bpw EXL2 Parameters and Internals

Use Cases 
Limitations:
This model is highly compliant and may respond to unethical requests.
Considerations:
You are advised to implement your own alignment layer before exposing the model as a service.
Supported Languages 
en (English)
Training Details 
Data Sources:
cognitivecomputations/dolphin, jondurbin/airoboros-2.2.1, cognitivecomputations/dolphin-coder, teknium/openhermes, ise-uiuc/Magicoder-OSS-Instruct-75K, ise-uiuc/Magicoder-Evol-Instruct-110K, LDJnr/Capybara
Methodology:
The model was retrained with some Mixtral-specific fixes in transformers library, and with gate layer unfrozen.
Context Length:
16000
Training Time:
3 days
Hardware Used:
4x A100s using qLoRA and Axolotl
Safety Evaluation 
Methodologies:
Filtering the dataset to remove alignment and bias
Ethical Considerations:
The model is uncensored and highly compliant to any requests, including possibly unethical ones.
Input Output 
Input Format:
ChatML prompt format
Release Notes 
Version:
2.7
Notes:
Retrained with Mixtral-specific fixes, with gate layer unfrozen.
Version:
2.6
Notes:
Fixed training configuration issue, added samantha-based empathy data, replaced synthia and pure-dove with Capybara.
LLM NameDolphin 2.7 Mixtral 8x7b 8bpw EXL2
Repository 🤗https://huggingface.co/Kearm/dolphin-2.7-mixtral-8x7b-8bpw-exl2 
Required VRAM46.8 GB
Updated2026-04-11
MaintainerKearm
Model Typemixtral
Instruction-BasedYes
Model Files  8.6 GB: 1-of-6   8.6 GB: 2-of-6   8.6 GB: 3-of-6   8.6 GB: 4-of-6   8.6 GB: 5-of-6   3.8 GB: 6-of-6
Supported Languagesen
Quantization Typeexl2
Model ArchitectureMixtralForCausalLM
Licenseapache-2.0
Context Length32768
Model Max Length32768
Transformers Version4.37.0.dev0
Tokenizer ClassLlamaTokenizer
Padding Token</s>
Vocabulary Size32002
Torch Data Typebfloat16

Best Alternatives to Dolphin 2.7 Mixtral 8x7b 8bpw EXL2

Best Alternatives
Context / RAM
Downloads
Likes
...M 2 8x22B Beige 5.0bpw H6 EXL264K / 88.5 GB110
...M 2 8x22B Beige 2.4bpw H6 EXL264K / 42.7 GB60
...M 2 8x22B Beige 3.0bpw H6 EXL264K / 53.2 GB60
...M 2 8x22B Beige 4.0bpw H6 EXL264K / 70.8 GB50
...B Instruct V0.1 8.0bpw H8 EXL264K / 120.2 GB41
...8x22b Instruct Oh EXL2 2.25bpw64K / 40.1 GB21
...eryTour V2 8x7B 4.5bpw H6 EXL232K / 26.5 GB42
...it MoE 2bitgs8 Metaoffload HQQ32K / 24.1 GB2819
... 4bit MoE 3bit Metaoffload HQQ32K / 22.4 GB213
... 4bit MoE 2bit Metaoffload HQQ32K / 18.3 GB1716
Note: green Score (e.g. "73.2") means that the model is better than Kearm/dolphin-2.7-mixtral-8x7b-8bpw-exl2.

Rank the Dolphin 2.7 Mixtral 8x7b 8bpw EXL2 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! 🌟

Instruction Following and Task Automation  
Factuality and Completeness of Knowledge  
Censorship and Alignment  
Data Analysis and Insight Generation  
Text Generation  
Text Summarization and Feature Extraction  
Code Generation  
Multi-Language Support and Translation  

What open-source LLMs or SLMs are you in search of? 52721 in total.

Email us: info@extractum.io. Our Privacy Policy  | Terms and Conditions  | Suggest an improvement.
Our Social Media →  
Original data from HuggingFace, OpenCompass and various public git repos.
Check out Ag3ntum — our secure, self-hosted AI agent for server management.
Release v20260328a
  Support LLM Explorer