Rhea 72B V0.5 4bit 128g 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)
Rhea 72B V0.5 4bit 128g Parameters and Internals
| LLM Name | Rhea 72B V0.5 4bit 128g |
| Repository 🤗 | https://huggingface.co/titan087/Rhea-72b-v0.5-4bit-128g |
| Model Size | 72b |
| Required VRAM | 41.3 GB |
| Updated | 2024-09-18 |
| Maintainer | titan087 |
| Model Type | llama |
| Model Files | |
| Supported Languages | en |
| GPTQ Quantization | Yes |
| Quantization Type | gptq|4bit |
| Model Architecture | LlamaForCausalLM |
| License | apache-2.0 |
| Context Length | 32768 |
| Model Max Length | 32768 |
| Transformers Version | 4.39.0.dev0 |
| Tokenizer Class | GPT2Tokenizer |
| Vocabulary Size | 152064 |
| Torch Data Type | float16 |
Best Alternatives to Rhea 72B V0.5 4bit 128g
Best Alternatives |
Context / RAM |
Downloads |
Likes |
|---|---|---|---|
| Smaug 72B V0.1 GPTQ | 32K / 41.3 GB | 50 | 8 |
| Smaug 72B V0.1 GPTQ | 32K / 41.3 GB | 23 | 2 |
| MoMo 72B Lora 1.8.7 DPO GPTQ | 32K / 41.3 GB | 23 | 7 |
| Smaug 72B V0.1 2.4bpw H6 EXL2 | 32K / 24.5 GB | 30 | 1 |
| Smaug 72B V0.1 | 32K / 144.5 GB | 3254 | 463 |
| TW3 JRGL V2 | 32K / 79.7 GB | 3134 | 0 |
| Le Triomphant ECE TW3 | 32K / 79.7 GB | 3129 | 3 |
| ECE TW3 JRGL V5 | 32K / 159.6 GB | 660 | 0 |
| MoMo 72B Lora 1.8.7 DPO | 32K / 208.5 GB | 769 | 68 |
| Rhea 72B V0.5 | 32K / 144.5 GB | 3247 | 132 |
Note: green Score (e.g. "73.2") means that the model is better than titan087/Rhea-72b-v0.5-4bit-128g .
Rank the Rhea 72B V0.5 4bit 128g 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? 36026 in total.
Original data from HuggingFace, OpenCompass and various public git repos.
Release v2024072803