We opensource our Aquila2 series, now including Aquila2, the base language models, namely Aquila2-7B and Aquila2-34B, as well as AquilaChat2, the chat models, namely AquilaChat2-7B and AquilaChat2-34B, as well as the long-text chat models, namely AquilaChat2-7B-16k and AquilaChat2-34B-16k
我们开源了我们的 Aquila2 系列，现在包括基础语言模型 Aquila2-7B 和 Aquila2-34B ，对话模型 AquilaChat2-7B 和 AquilaChat2-34B，长文本对话模型AquilaChat2-7B-16k 和 AquilaChat2-34B-16k
2023.10.25 ? AquilaChat2-34B v1.2 is based on the previous AquilaChat2-34B. The AquilaChat2-34B model is close to or exceeds the level of GPT3.5 in the subjective evaluation of 8 secondary ability dimensions.
The additional details of the Aquila model will be presented in the official technical report. Please stay tuned for updates on official channels.
基于AquilaChat2-34B初始版本的开发经验，我们对AquilaChat2-34B进行了全面升级并发布1.2版本。评测结果显示， AquilaChat2-34B-V1.2 模型在主观评测的8个二级能力维度上，均接近或超过 GPT3.5 水平。
悟道 · 天鹰 Aquila 模型的更多细节将在官方技术报告中呈现。请关注官方渠道更新。
Quick Start 快速开始使用 AquilaChat2-34B（Chat model）
1. Inference 推理
from modelscope import AutoModelForCausalLM, AutoTokenizer, snapshot_download
from predict import predict
import torch
# Note: The default behavior now has injection attack prevention off.
device = torch.device("cuda:0")
model_dir = snapshot_download("BAAI/AquilaChat2-34B")
tokenizer = AutoTokenizer.from_pretrained(model_dir, trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained(model_dir, device_map="auto", trust_remote_code=True).eval()
model.to(device)
text = "请给出10个要到北京旅游的理由。"
from predict import predict
out = predict(model, text, tokenizer=tokenizer, max_gen_len=200, top_p=0.9,
seed=123, topk=15, temperature=1.0, sft=True, device=device,
model_name="AquilaChat2-34B")
print(out)
证书/License
Aquila2 series open-source model is licensed under BAAI Aquila Model Licence Agreement
Aquila2系列开源模型使用 智源Aquila系列模型许可协议
评论