Llama 3 70B Quantised 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)
Llama 3 70B Quantised Parameters and Internals
| LLM Name | Llama 3 70B Quantised |
| Repository 🤗 | https://huggingface.co/farhan-ahmad/llama-3-70b-quantised |
| Model Size | 70b |
| Required VRAM | 48.7 GB |
| Updated | 2024-09-19 |
| Maintainer | farhan-ahmad |
| Model Type | llama |
| Model Files | |
| GGUF Quantization | Yes |
| Quantization Type | gguf |
| Model Architecture | LlamaForCausalLM |
| License | mit |
| Context Length | 8192 |
| Model Max Length | 8192 |
| Transformers Version | 4.40.0.dev0 |
| Vocabulary Size | 128256 |
| Torch Data Type | bfloat16 |
Best Alternatives to Llama 3 70B Quantised
Best Alternatives |
Context / RAM |
Downloads |
Likes |
|---|---|---|---|
| Reflection Llama 3.1 70B GGUF | 128K / 26.4 GB | 402 | 3 |
| Reflection Llama 3.1 70B Bf16 | 128K / 141.9 GB | 283 | 6 |
| Midnight Miqu 70B V1.0 GGUF | 31K / 29.9 GB | 83 | 4 |
| ...qu 1 70B 24GB VRAM IQ2 XS SOTA | 31K / 20.3 GB | 63 | 0 |
| ...ma3 70B Chinese Chat GGUF 4bit | 8K / 40 GB | 1192 | 17 |
| ...3 Mega Dolphin 2.9.1 120b GGUF | 8K / 18.4 GB | 24 | 1 |
| Meta Llama 3 70B Instruct | 8K / 40.3 GB | 6 | 1 |
| CodeLlama 70B Instruct Hf GGUF | 4K / 25.5 GB | 169 | 2 |
| Openthaigpt 1.0.0 70B Chat | 2K / 138.4 GB | 169 | 11 |
| Maywell Kiqu 70B GGUF | 0K / 14.5 GB | 73 | 3 |
Note: green Score (e.g. "73.2") means that the model is better than farhan-ahmad/llama-3-70b-quantised .
Rank the Llama 3 70B Quantised 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? 36073 in total.
Original data from HuggingFace, OpenCompass and various public git repos.
Release v2024072803