Autotrain compatible Base model:finetune:moreh/momo... Base model:moreh/momo-72b-lora... Endpoints compatible Exl2 Llama Lora Quantized Region:us Safetensors Sharded Tensorflow
Smaug 72B V0.1 2.4bpw H6 EXL2 Benchmarks
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)
Smaug 72B V0.1 2.4bpw H6 EXL2 Parameters and Internals
| LLM Name | Smaug 72B V0.1 2.4bpw H6 EXL2 |
| Repository 🤗 | https://huggingface.co/LoneStriker/Smaug-72B-v0.1-2.4bpw-h6-exl2 |
| Base Model(s) | |
| Model Size | 72b |
| Required VRAM | 24.5 GB |
| Updated | 2024-09-16 |
| Maintainer | LoneStriker |
| Model Type | llama |
| Model Files | |
| Quantization Type | exl2 |
| Model Architecture | LlamaForCausalLM |
| License | other |
| Context Length | 32768 |
| Model Max Length | 32768 |
| Transformers Version | 4.36.2 |
| Tokenizer Class | GPT2Tokenizer |
| Vocabulary Size | 152064 |
| LoRA Model | Yes |
| Torch Data Type | bfloat16 |
Best Alternatives to Smaug 72B V0.1 2.4bpw H6 EXL2
Best Alternatives |
Context / RAM |
Downloads |
Likes |
|---|---|---|---|
| Smaug 72B V0.1 | 32K / 144.5 GB | 3235 | 463 |
| TW3 JRGL V2 | 32K / 79.7 GB | 3104 | 0 |
| Le Triomphant ECE TW3 | 32K / 79.7 GB | 3108 | 3 |
| ECE TW3 JRGL V5 | 32K / 159.6 GB | 655 | 0 |
| MoMo 72B Lora 1.8.7 DPO | 32K / 208.5 GB | 760 | 68 |
| Rhea 72B V0.5 | 32K / 144.5 GB | 3209 | 132 |
| JuliusCesar 72B BeyonderV.0 | 32K / 74.2 GB | 59 | 0 |
| ECE TW3 JRGL V3 | 32K / 77.8 GB | 2473 | 0 |
| ECE TW3 JRGL V4 | 32K / 77.8 GB | 2446 | 0 |
| Caigun Model 72B KGI | 32K / 144.6 GB | 5 | 0 |
Note: green Score (e.g. "73.2") means that the model is better than LoneStriker/Smaug-72B-v0.1-2.4bpw-h6-exl2 .
Rank the Smaug 72B V0.1 2.4bpw H6 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! 🌟
What open-source LLMs or SLMs are you in search of? 35941 in total.
Original data from HuggingFace, OpenCompass and various public git repos.
Release v2024072803