llm 1.3.8

A Rust library unifying multiple LLM backends.
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

use serde::{Deserialize, Serialize};

/// Request payload for chat completion API endpoint
#[derive(Deserialize)]
pub struct ChatRequest {
    /// List of messages in the conversation
    pub messages: Option<Vec<Message>>,
    /// Model identifier in format "provider:model_name"
    pub model: Option<String>,
    /// Optional chain steps for multi-step processing
    #[serde(default)]
    pub steps: Vec<ChainStepRequest>,
    /// Optional response transformation function
    #[serde(default)]
    pub response_transform: Option<String>,
    /// Optional temperature parameter
    #[serde(default)]
    pub temperature: Option<f32>,
    /// Optional max tokens parameter
    #[serde(default)]
    pub max_tokens: Option<u32>,
}

/// Chain step configuration for multi-step processing
#[derive(Deserialize)]
pub struct ChainStepRequest {
    /// Provider ID for this step
    pub provider_id: String,
    /// Step identifier
    pub id: String,
    /// Template with variable substitution
    pub template: String,
    /// Optional temperature parameter
    #[serde(default)]
    pub temperature: Option<f32>,
    /// Optional max tokens parameter
    #[serde(default)]
    pub max_tokens: Option<u32>,
    /// Optional response transformation function
    #[serde(default)]
    pub response_transform: Option<String>,
}

/// Single message in a chat conversation
#[derive(Deserialize, Serialize)]
pub struct Message {
    /// Role of the message sender ("user" or "assistant")
    pub role: String,
    /// Content of the message
    pub content: String,
}

/// Response payload from chat completion API endpoint
#[derive(Serialize)]
pub struct ChatResponse {
    /// Unique identifier for this completion
    pub id: String,
    /// Object type identifier
    pub object: String,
    /// Unix timestamp when response was created
    pub created: u64,
    /// Name of the model that generated the completion
    pub model: String,
    /// List of completion choices generated
    pub choices: Vec<Choice>,
}

/// Single completion choice in a chat response
#[derive(Serialize)]
pub struct Choice {
    /// Index of this choice in the list
    pub index: usize,
    /// Generated message for this choice
    pub message: Message,
    /// Reason why the model stopped generating
    pub finish_reason: String,
}