LLM EXPLORER 60,783 MODELS INDEXED

Dolphin 2.9.1 Llama 3 70B 6.0bpw EXL2 by bigstorm

By bigstorm · 5 downloads
HF Repo ↗ Copy URL

Dolphin 2.9.1 Llama 3 70B 6.0bpw EXL2 is an open-source language model by bigstorm. Features: 70b LLM, VRAM: 54.1GB, Context: 8K, License: llama3, Quantized, LLM Explorer Score: 0.13.

  6-bit   Axolotl Base model:meta-llama/meta-lla... Base model:quantized:meta-llam...   Conversational Dataset:cognitivecomputations/... Dataset:cognitivecomputations/... Dataset:cognitivecomputations/...   Dataset:internlm/agent-flan Dataset:locutusque/function-ca... Dataset:m-a-p/codefeedback-fil... Dataset:microsoft/orca-math-wo...   Dataset:teknium/openhermes-2.5   Endpoints compatible   Exl2   Generated from trainer   Llama   Quantized   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/bigstorm/dolphin-2.9.1-llama-3-70b-6.0bpw-exl2 

Dolphin 2.9.1 Llama 3 70B 6.0bpw EXL2 Parameters and Internals

Model Type 
Instruction, Conversational, Coding
Use Cases 
Areas:
Instruction, Conversation, Coding
Primary Use Cases:
Function calling
Limitations:
Highly compliant with any requests, even unethical ones.
Considerations:
Implement your own alignment layer before exposing the model as a service.
Additional Notes 
Dolphin is licensed according to Meta's Llama license. Trained on data generated from GPT4, among others.
Training Details 
Data Sources:
GPT4, ShareGPT, ChatML
Methodology:
FFT, with ChatML prompt template format
Context Length:
32000
Training Time:
3 days
Hardware Used:
8x H100 from Crusoe Cloud
LLM NameDolphin 2.9.1 Llama 3 70B 6.0bpw EXL2
Repository 🤗https://huggingface.co/bigstorm/dolphin-2.9.1-llama-3-70b-6.0bpw-exl2 
Base Model(s)  Meta Llama 3 70B   meta-llama/Meta-Llama-3-70B
Model Size70b
Required VRAM54.1 GB
Updated2026-07-27
Maintainerbigstorm
Model Typellama
Model Files  8.6 GB: 1-of-7   8.4 GB: 2-of-7   8.6 GB: 3-of-7   8.5 GB: 4-of-7   8.4 GB: 5-of-7   8.5 GB: 6-of-7   3.1 GB: 7-of-7
Quantization Typeexl2
Model ArchitectureLlamaForCausalLM
Licensellama3
Context Length8192
Model Max Length8192
Transformers Version4.40.2
Tokenizer ClassPreTrainedTokenizerFast
Padding Token<|end_of_text|>
Vocabulary Size128258
Torch Data Typebfloat16

Best Alternatives to Dolphin 2.9.1 Llama 3 70B 6.0bpw EXL2

Best Alternatives
Context / RAM
Downloads
Likes
...B Instruct Gradient 1048K 8bit1024K / 75 GB415
...B Instruct Gradient 1048K 4bit1024K / 39.7 GB253
...B Instruct Gradient 1048K 2bit1024K / 21.9 GB252
...0B Instruct Gradient 262K 4bit256K / 39.7 GB253
...0B Instruct Gradient 262K 8bit256K / 75 GB262
...0B Instruct Gradient 262K 2bit256K / 21.9 GB201
... Gradient 262K 2.25bpw H6 EXL2256K / 22.2 GB80
...t Gradient 262K 4.0bpw H6 EXL2256K / 37.2 GB71
...t Gradient 262K 2.4bpw H6 EXL2256K / 23.5 GB60
...t Gradient 262K 3.5bpw H6 EXL2256K / 32.9 GB60
Note: green Score (e.g. "73.2") means that the model is better than bigstorm/dolphin-2.9.1-llama-3-70b-6.0bpw-exl2.