Total runs:
64.3M
Run Growth:
-6.6M
Growth Rate:
-10.32%
Sponsored by
YouArt
BAAI / AquilaChat2-34B
Model's Last Updated: July 13 2026
text-generation
Introduction of AquilaChat2-34B
Model Details of AquilaChat2-34B
English | 简体中文
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
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.
Note
We have discovered a data leakage problem with the GSM8K test data in the pre-training task dataset. Therefore, the evaluation results of GSM8K have been removed from the evaluation results.
Upon thorough investigation and analysis, it was found that the data leakage occurred in the mathematical dataset A (over 2 million samples), recommended by a team we have collaborated with multiple times. This dataset includes the untreated GSM8K test set (1319 samples). The team only performed routine de-duplication and quality checks but did not conduct an extra filtering check for the presence of the GSM8K test data, resulting in this oversight.
Our team has always strictly adhered to the principle that training data should not include test data. Taking this lesson from the error caused by not thoroughly checking the source of external data, we have investigated all 2 trillion tokens of data for various test datasets, including WTM22（en-zh), CLUEWSC, Winograd, HellaSwag, OpenBookQA, PIQA, ARC-e, BUSTSM, BoolQ, TruthfulQA, RAFT, ChID, EPRSTMT, TNEWS, OCNLI, SEM-Chinese, MMLU, C-Eval, CMMLU, CSL and HumanEval.
Quick Start AquilaChat2-34B（Chat model）
1. Inference
from transformers import AutoTokenizer, AutoModelForCausalLM
from transformers import BitsAndBytesConfig
import torch
device = torch.device("cuda:0")
model_info = "BAAI/AquilaChat2-34B"
tokenizer = AutoTokenizer.from_pretrained(model_info, trust_remote_code=True)
quantization_config=BitsAndBytesConfig(
load_in_4bit=True,
bnb_4bit_use_double_quant=True,
bnb_4bit_quant_type="nf4",
bnb_4bit_compute_dtype=torch.bfloat16,
)
model = AutoModelForCausalLM.from_pretrained(model_info, trust_remote_code=True, torch_dtype=torch.bfloat16,
# quantization_config=quantization_config, # Uncomment this line for 4bit quantization
)
model.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)
License
Aquila2 series open-source model is licensed under BAAI Aquila Model Licence Agreement
Runs of BAAI AquilaChat2-34B on huggingface.co
157
Total runs
4
24-hour runs
4
3-day runs
6
7-day runs
29
30-day runs
More Information About AquilaChat2-34B huggingface.co Model
Url of AquilaChat2-34B
AquilaChat2-34B huggingface.co Url
Provider of AquilaChat2-34B huggingface.co
Other API from BAAI
Updated:February 22 2024
Total runs:
18.2M
Run Growth:
-19.5K
Growth Rate:
-0.11%
Total runs:
12.8M
Run Growth:
622.5K
Growth Rate:
4.86%
Total runs:
10.6M
Run Growth:
-274.5K
Growth Rate:
-2.59%
Total runs:
5.0M
Run Growth:
289.8K
Growth Rate:
5.83%
Total runs:
3.7M
Run Growth:
-294.3K
Growth Rate:
-7.87%
Total runs:
1.2M
Run Growth:
-263.2K
Growth Rate:
-21.69%
Total runs:
795.3K
Run Growth:
-726.2K
Growth Rate:
-91.32%
Total runs:
237.1K
Run Growth:
143.6K
Growth Rate:
59.67%
Total runs:
228.9K
Run Growth:
-1.3M
Growth Rate:
-582.76%
Total runs:
189.7K
Run Growth:
-741.6K
Growth Rate:
-390.97%
Total runs:
163.9K
Run Growth:
5.7K
Growth Rate:
3.50%
Total runs:
59.2K
Run Growth:
-116.7K
Growth Rate:
-197.13%
Total runs:
16.7K
Run Growth:
1.1K
Growth Rate:
6.79%
Total runs:
11.1K
Run Growth:
-15.2K
Growth Rate:
-136.62%
Total runs:
8.7K
Run Growth:
387
Growth Rate:
4.47%
Total runs:
8.6K
Run Growth:
538
Growth Rate:
6.25%
Total runs:
4.8K
Run Growth:
-1.3K
Growth Rate:
-26.89%
Total runs:
3.4K
Run Growth:
2.4K
Growth Rate:
72.02%
Total runs:
2.7K
Run Growth:
1.9K
Growth Rate:
70.21%
Total runs:
2.1K
Run Growth:
-625
Growth Rate:
-30.08%
Total runs:
1.2K
Run Growth:
-1.6K
Growth Rate:
-139.41%
Total runs:
845
Run Growth:
-87
Growth Rate:
-10.30%
Total runs:
664
Run Growth:
0
Growth Rate:
0.00%
Total runs:
644
Run Growth:
0
Growth Rate:
0.00%
Total runs:
609
Run Growth:
609
Growth Rate:
100.00%
Total runs:
603
Run Growth:
-856
Growth Rate:
-141.96%
Total runs:
564
Run Growth:
-652
Growth Rate:
-115.60%
Total runs:
327
Run Growth:
-71
Growth Rate:
-21.71%
Total runs:
285
Run Growth:
-3.8K
Growth Rate:
-1338.60%
Total runs:
179
Run Growth:
53
Growth Rate:
29.61%
Total runs:
168
Run Growth:
61
Growth Rate:
36.31%
Total runs:
163
Run Growth:
72
Growth Rate:
44.17%
Total runs:
162
Run Growth:
-527
Growth Rate:
-325.31%
Total runs:
146
Run Growth:
52
Growth Rate:
35.62%
Total runs:
142
Run Growth:
12
Growth Rate:
8.45%