LLM EXPLORER 59,420 MODELS INDEXED

Dolphin 2.9.2 Qwen2 72B 6.0bpw EXL2 by bigstorm

By bigstorm · 5 downloads
HF Repo ↗ Copy URL

Dolphin 2.9.2 Qwen2 72B 6.0bpw EXL2 is an open-source language model by bigstorm. Features: 72b LLM, VRAM: 56.1GB, Context: 128K, License: other, Quantized, LLM Explorer Score: 0.13.

  6-bit   Axolotl Base model:quantized:qwen/qwen...   Base model:qwen/qwen2-72b   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   Exllamasv2   Generated from trainer   Quantized   Qwen2   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/bigstorm/dolphin-2.9.2-qwen2-72b-6.0bpw-exl2 

Dolphin 2.9.2 Qwen2 72B 6.0bpw EXL2 Parameters and Internals

Model Type 
AutoModelForCausalLM
Additional Notes 
Dolphin-2.9.2 has a variety of instruction, conversational, and coding skills. It also has initial agentic abilities and supports function calling. Dolphin is licensed according to Qwen's tongyi-qianwen license. We grant permission for any use, including commercial, that falls within accordance with said license.
Training Details 
Data Sources:
GPT4, cognitivecomputations/Dolphin-2.9, teknium/OpenHermes-2.5, m-a-p/CodeFeedback-Filtered-Instruction, cognitivecomputations/dolphin-coder, cognitivecomputations/samantha-data, microsoft/orca-math-word-problems-200k, Locutusque/function-calling-chatml, internlm/Agent-FLAN
Methodology:
FFT on parameters selected by Laser Scanner, using ChatML prompt template format
Context Length:
8192
Hardware Used:
Crusoe Cloud 8xH100 node
Safety Evaluation 
Ethical Considerations:
You are advised to implement your own alignment layer before exposing the model as a service. It will be highly compliant with any requests, even unethical ones.
LLM NameDolphin 2.9.2 Qwen2 72B 6.0bpw EXL2
Repository 🤗https://huggingface.co/bigstorm/dolphin-2.9.2-qwen2-72b-6.0bpw-exl2 
Base Model(s)  Qwen2 72B   Qwen/Qwen2-72B
Model Size72b
Required VRAM56.1 GB
Updated2026-08-02
Maintainerbigstorm
Model Typeqwen2
Model Files  8.5 GB: 1-of-7   8.4 GB: 2-of-7   8.5 GB: 3-of-7   8.5 GB: 4-of-7   8.5 GB: 5-of-7   8.4 GB: 6-of-7   5.3 GB: 7-of-7
Quantization Typeexl2
Model ArchitectureQwen2ForCausalLM
Licenseother
Context Length131072
Model Max Length131072
Transformers Version4.40.2
Tokenizer ClassQwen2Tokenizer
Padding Token<|endoftext|>
Vocabulary Size152064
Torch Data Typebfloat16
Errorsreplace

Best Alternatives to Dolphin 2.9.2 Qwen2 72B 6.0bpw EXL2

Best Alternatives
Context / RAM
Downloads
Likes
Kimi Dev 72B 4bit DWQ128K / 40.9 GB446521
Kimi Dev 72B 8bit128K / 77.1 GB3612
Kimi Dev 72B 5bit128K / 50.1 GB3682
Qwen2.5 72B Bnb 4bit128K / 41.4 GB1531
Qwen2 72B Bnb 4bit128K / 41.2 GB994
...Qwen2 72B 4 0bpw H6 EXL2 Pippa128K / 38.6 GB71
...phin 292 Qwen2 72b EXL2 4 0bpw128K / 38.6 GB41
Qwen2 72B 4bit128K / 40.9 GB510
...n 2.9.2 Qwen72b 8.0bpw H8 EXL2128K / 66.8 GB32
...B Instruct 2.0bpw H Novel EXL2125K / 23 GB41
Note: green Score (e.g. "73.2") means that the model is better than bigstorm/dolphin-2.9.2-qwen2-72b-6.0bpw-exl2.