aha 0.2.5

aha model inference library, now supports Qwen(2.5VL/3/3VL/3.5/ASR/3Embedding/3Reranker), MiniCPM4, VoxCPM/1.5, DeepSeek-OCR/2, Hunyuan-OCR, PaddleOCR-VL/1.5, RMBG2.0, GLM(ASR-Nano-2512/OCR), Fun-ASR-Nano-2512, LFM(2/2.5/2VL/2.5VL)
Documentation 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142

use std::{pin::pin, time::Instant};

use aha::models::{GenerateModel, paddleocr_vl::generate::PaddleOCRVLGenerateModel};
use aha::params::chat::ChatCompletionParameters;
use anyhow::Result;
use rocket::futures::StreamExt;

#[test]
fn paddleocr_vl_1_5_generate() -> Result<()> {
    // RUST_BACKTRACE=1 cargo test -F cuda --test test_paddleocr_vl paddleocr_vl_1_5_generate -r -- --nocapture
    let message = r#"
    {
        "model": "paddleocr_vl1.5",
        "messages": [
            {
                "role": "user",
                "content": [ 
                    {
                        "type": "image",
                        "image_url": 
                        {
                            "url": "file://./assets/img/ocr_test1.png"
                        }
                    },              
                    {
                        "type": "text", 
                        "text": "Spotting:"
                    }
                ]
            }
        ],
        "stream": false
    }
    "#;
    let save_dir =
        aha::utils::get_default_save_dir().ok_or(anyhow::anyhow!("Failed to get save dir"))?;
    let model_path = format!("{}/PaddlePaddle/PaddleOCR-VL-1.5/", save_dir);
    let mes: ChatCompletionParameters = serde_json::from_str(message)?;
    let i_start = Instant::now();
    let mut model = PaddleOCRVLGenerateModel::init(&model_path, None, None)?;
    let i_duration = i_start.elapsed();
    println!("Time elapsed in load model is: {:?}", i_duration);
    let i_start = Instant::now();
    let res = model.generate(mes)?;
    let i_duration = i_start.elapsed();
    println!("generate: \n {:?}", res);
    if let Some(usage) = &res.usage {
        let num_token = usage.total_tokens;
        let duration_secs = i_duration.as_secs_f64();
        let tps = num_token as f64 / duration_secs;
        println!("Tokens per second (TPS): {:.2}", tps);
    }
    println!("Time elapsed in generate is: {:?}", i_duration);
    Ok(())
}

#[test]
fn paddleocr_vl_generate() -> Result<()> {
    // RUST_BACKTRACE=1 cargo test -F cuda --test test_paddleocr_vl paddleocr_vl_generate -r -- --nocapture
    let message = r#"
    {
        "model": "paddleocr_vl",
        "messages": [
            {
                "role": "user",
                "content": [ 
                    {
                        "type": "image",
                        "image_url": 
                        {
                            "url": "file://./assets/img/ocr_test1.png"
                        }
                    },              
                    {
                        "type": "text", 
                        "text": "OCR:"
                    }
                ]
            }
        ],
        "stream": false
    }
    "#;
    let save_dir =
        aha::utils::get_default_save_dir().ok_or(anyhow::anyhow!("Failed to get save dir"))?;
    let model_path = format!("{}/PaddlePaddle/PaddleOCR-VL/", save_dir);
    let mes: ChatCompletionParameters = serde_json::from_str(message)?;
    let i_start = Instant::now();
    let mut model = PaddleOCRVLGenerateModel::init(&model_path, None, None)?;
    let i_duration = i_start.elapsed();
    println!("Time elapsed in load model is: {:?}", i_duration);
    let res = model.generate(mes)?;
    println!("generate: \n {:?}", res);
    if let Some(usage) = &res.usage {
        println!("usage: \n {:?}", usage);
    }
    Ok(())
}

#[tokio::test]
async fn paddleocr_vl_stream() -> Result<()> {
    // test with cuda: RUST_BACKTRACE=1 cargo test -F cuda paddleocr_vl_stream -r -- --nocapture

    let message = r#"
    {
        "model": "paddleocr_vl",
        "messages": [
            {
                "role": "user",
                "content": [ 
                    {
                        "type": "image",
                        "image_url": 
                        {
                            "url": "https://www.qqxiuzi.cn/zh/shouxie-shufa/welcome.png"
                        }
                    },              
                    {
                        "type": "text", 
                        "text": "OCR:"
                    }
                ]
            }
        ]
    }
    "#;
    let save_dir =
        aha::utils::get_default_save_dir().ok_or(anyhow::anyhow!("Failed to get save dir"))?;
    let model_path = format!("{}/PaddlePaddle/PaddleOCR-VL/", save_dir);
    let mes: ChatCompletionParameters = serde_json::from_str(message)?;
    let i_start = Instant::now();
    let mut model = PaddleOCRVLGenerateModel::init(&model_path, None, None)?;
    let i_duration = i_start.elapsed();
    println!("Time elapsed in load model is: {:?}", i_duration);
    let mut stream = pin!(model.generate_stream(mes)?);
    while let Some(item) = stream.next().await {
        println!("generate: \n {:?}", item);
    }
    let i_duration = i_start.elapsed();
    println!("Time elapsed in generate is: {:?}", i_duration);
    Ok(())
}