litellm-rs 0.4.16

A high-performance AI Gateway written in Rust, providing OpenAI-compatible APIs with intelligent routing, load balancing, and enterprise features
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

//! NLP Cloud Model Information

use std::collections::HashMap;
use std::sync::LazyLock;

#[derive(Debug, Clone)]
pub struct ModelInfo {
    pub model_id: &'static str,
    pub display_name: &'static str,
    pub max_context_length: u32,
    pub max_output_length: u32,
    pub supports_tools: bool,
    pub supports_multimodal: bool,
    pub input_cost_per_million: f64,
    pub output_cost_per_million: f64,
}

static MODEL_CONFIGS: LazyLock<HashMap<&'static str, ModelInfo>> = LazyLock::new(|| {
    let mut configs = HashMap::new();

    configs.insert(
        "finetuned-llama-3-70b",
        ModelInfo {
            model_id: "finetuned-llama-3-70b",
            display_name: "Fine-tuned Llama 3 70B",
            max_context_length: 8192,
            max_output_length: 4096,
            supports_tools: false,
            supports_multimodal: false,
            input_cost_per_million: 1.00,
            output_cost_per_million: 1.50,
        },
    );

    configs.insert(
        "dolphin-mixtral-8x7b",
        ModelInfo {
            model_id: "dolphin-mixtral-8x7b",
            display_name: "Dolphin Mixtral 8x7B",
            max_context_length: 32768,
            max_output_length: 8192,
            supports_tools: false,
            supports_multimodal: false,
            input_cost_per_million: 0.80,
            output_cost_per_million: 1.00,
        },
    );

    configs.insert(
        "chatdolphin",
        ModelInfo {
            model_id: "chatdolphin",
            display_name: "ChatDolphin",
            max_context_length: 16384,
            max_output_length: 4096,
            supports_tools: false,
            supports_multimodal: false,
            input_cost_per_million: 0.50,
            output_cost_per_million: 0.70,
        },
    );

    configs
});

pub fn get_model_info(model_id: &str) -> Option<&'static ModelInfo> {
    MODEL_CONFIGS.get(model_id)
}

pub fn get_available_models() -> Vec<&'static str> {
    MODEL_CONFIGS.keys().copied().collect()
}

#[cfg(test)]
mod tests {
    use super::*;

    #[test]
    fn test_get_model_info() {
        let info = get_model_info("finetuned-llama-3-70b");
        assert!(info.is_some());
    }
}