metadata
license: apache-2.0
language:
- zh
- en
Llama-3-Chinese-8B-Instruct-GGUF
Warning: llama.cpp has breaking changes on Llama-3 pre-tokenizer, which significantly affect performance. We will update GGUF mdoels in the next few hours.
This repository contains Llama-3-Chinese-8B-Instruct-GGUF (llama.cpp/ollama/tgw, etc. compatible), which is the quantized version of Llama-3-Chinese-8B-Instruct.
Note: this is an instruction (chat) model, which can be used for conversation, QA, etc.
Further details (performance, usage, etc.) should refer to GitHub project page: https://github.com/ymcui/Chinese-LLaMA-Alpaca-3
Performance
Metric: PPL, lower is better
| Quant | Size | PPL | PPL (-im) |
|---|---|---|---|
| Q2_K | 2.96 GB | 10.3918 +/- 0.13288 | 9.1722 +/- 0.11502 |
| Q3_K | 3.74 GB | 6.3018 +/- 0.07849 | 6.1901 +/- 0.07734 |
| Q4_0 | 4.34 GB | 6.0628 +/- 0.07501 | 5.9623 +/- 0.07444 |
| Q4_K | 4.58 GB | 5.9066 +/- 0.07419 | 5.8847 +/- 0.07406 |
| Q5_0 | 5.21 GB | 5.8562 +/- 0.07355 | 5.8032 +/- 0.07287 |
| Q5_K | 5.34 GB | 5.8062 +/- 0.07331 | 5.8058 +/- 0.07329 |
| Q6_K | 6.14 GB | 5.7757 +/- 0.07298 | 5.7745 +/- 0.07287 |
| Q8_0 | 7.95 GB | 5.7626 +/- 0.07272 | 5.7626 +/- 0.07272 |
| F16 | 14.97 GB | 5.7628 +/- 0.07275 | N/A |
Others
For full model, please see: https://huggingface.co/hfl/llama-3-chinese-8b-instruct
For LoRA-only model, please see: https://huggingface.co/hfl/llama-3-chinese-8b-instruct-lora
If you have questions/issues regarding this model, please submit an issue through https://github.com/ymcui/Chinese-LLaMA-Alpaca-3