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，Aquila2-34B 和Aquila2-70B-Expr，对话模型 AquilaChat2-7B ，AquilaChat2-34B和AquilaChat2-70B-Expr，长文本对话模型AquilaChat2-7B-16k 和 AquilaChat2-34B-16k
2023.10.25 ? AquilaChat2-34B-16K v1.2 is based on the previous AquilaChat2-34B-16K. The AquilaChat2-34B-16K-V1.2 has significantly improved long-text synthesis capabilities compared to the V1 version, approaching the level of GPT-3.5-16K. Additionally, the V1.2 version incorporates more conventional instruction fine-tuning corpora, enhancing its performance in non-long-text scenarios compared to the V1 version.
The additional details of the Aquila model will be presented in the official technical report. Please stay tuned for updates on official channels.
2023.10.25 ? 基于AquilaChat2-34B-16K初始版本的开发经验，我们对AquilaChat2-34B-16K进行了全面升级并发布1.2版本。 其中AquilaChat2-34B-16K-V1.2相较于V1版本在长文本综合能力上有明显提升，接近GPT-3.5-16K。同时V1.2版本应用了更多的常规指令微调语料， 使其在非长文本场景下的性能也优于V1版本。
2023.12.05 ? 开源Aquila2-70B-Expr和AquilaChat2-70B-Expr，补齐了70B参数规模的中英双语大模型缺口，仅经过1.2T数据训练，依然表现出优秀潜力，可用于继续预训练及微调，为学界和业界的后续工作提供探索起点。
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 Aquila2-70B-Expr
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.
model_dir = snapshot_download("BAAI/Aquila2-70B-Expr")
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.eval()
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="Aquila2-70B-Expr")
print(out)
证书/License
Aquila2系列开源模型使用 (https://huggingface.co/BAAI/AquilaChat2-34B/blob/main/BAAI-Aquila-Model-License%20-Agreement.pdf)
评论