v3.0
版本发布时间: 2024-05-30 08:45:20
ymcui/Chinese-LLaMA-Alpaca-3最新发布版本:v3.0(2024-05-30 08:45:20)
本次发布Llama-3-Chinese-8B-Instruct-v3（指令模型），融合了v1、v2以及Meta原版Instruct模型，在中文任务上大幅超越v1/v2版，英文任务上与Meta原版保持持平，主观体验效果显著提升。
下载方式
| 模型名称 | 完整版 | LoRA版 | GGUF版 |
|---|---|---|---|
| Llama-3-Chinese-8B-Instruct-v3 (指令模型) |
[🤗Hugging Face] [🤖ModelScope] [wisemodel] |
N/A | [🤗Hugging Face] [🤖ModelScope] |
模型效果
客观任务效果请查阅：https://github.com/ymcui/Chinese-LLaMA-Alpaca-3?tab=readme-ov-file#模型效果
Llama-3-Chinese-8B-Instruct-v3已登陆大模型竞技场，当前胜率及Elo评分显著超越其他模型。
⚔️ 模型竞技场：http://llm-arena.ymcui.com/
What's Changed
- Add Open LLM Leaderboard results by @ymcui in https://github.com/ymcui/Chinese-LLaMA-Alpaca-3/pull/27
- Fix for termination issue by @akxlr in https://github.com/ymcui/Chinese-LLaMA-Alpaca-3/pull/38
- Add GPT-4o ruozhiba data & clean GPT-4t data by @ymcui in https://github.com/ymcui/Chinese-LLaMA-Alpaca-3/pull/45
- Update script with MPS backend & fix minor by @ymcui in https://github.com/ymcui/Chinese-LLaMA-Alpaca-3/pull/55
- Fix SFT training data format by @iMountTai in https://github.com/ymcui/Chinese-LLaMA-Alpaca-3/pull/60
New Contributors
- @akxlr made their first contribution in https://github.com/ymcui/Chinese-LLaMA-Alpaca-3/pull/38
Full Changelog: https://github.com/ymcui/Chinese-LLaMA-Alpaca-3/compare/v2.0...v3.0