llama-runner 2.3.2

A straightforward Rust library for running llama.cpp models locally on device
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

use crate::{
    GenericTextLmRequest, GenericVisionLmRequest, TextLmRunner, VisionLmRunner,
    error::GenericRunnerError, template::ChatTemplate,
};

pub trait TextLmRunnerExt<'s, 'req, Tmpl>
where
    Tmpl: ChatTemplate,
{
    fn get_lm_response(
        &'s self,
        request: GenericTextLmRequest<'req, Tmpl>,
    ) -> Result<String, GenericRunnerError<Tmpl::Error>>;
}

pub trait VisionLmRunnerExt<'s, 'req, Tmpl>
where
    Tmpl: ChatTemplate,
{
    fn get_vlm_response(
        &'s self,
        request: GenericVisionLmRequest<'req, Tmpl>,
    ) -> Result<String, GenericRunnerError<Tmpl::Error>>;
}

impl<'s, 'req, TextRunner, Tmpl> TextLmRunnerExt<'s, 'req, Tmpl> for TextRunner
where
    TextRunner: TextLmRunner<'s, 'req, Tmpl>,
    Tmpl: ChatTemplate,
{
    fn get_lm_response(
        &'s self,
        request: GenericTextLmRequest<'req, Tmpl>,
    ) -> Result<String, GenericRunnerError<Tmpl::Error>> {
        self.stream_lm_response(request)
            .collect::<Result<String, _>>()
    }
}

impl<'s, 'req, VisionRunner, Tmpl> VisionLmRunnerExt<'s, 'req, Tmpl> for VisionRunner
where
    VisionRunner: VisionLmRunner<'s, 'req, Tmpl>,
    Tmpl: ChatTemplate,
{
    fn get_vlm_response(
        &'s self,
        request: GenericVisionLmRequest<'req, Tmpl>,
    ) -> Result<String, GenericRunnerError<Tmpl::Error>> {
        self.stream_vlm_response(request)
            .collect::<Result<String, _>>()
    }
}