starweaver-model 0.8.0

Provider-neutral model protocol and wire adapters for Starweaver
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

use serde_json::Value;

use crate::{
    ModelError,
    adapter::ModelRequestParameters,
    message::{ModelMessage, ModelResponse},
    profile::{ModelProfile, ProtocolFamily},
    providers::{
        anthropic::AnthropicMessagesAdapter, bedrock::BedrockConverseAdapter,
        gemini::GeminiGenerateContentAdapter, openai_chat::OpenAiChatAdapter,
        openai_responses::OpenAiResponsesAdapter,
    },
    settings::{ModelSettings, ThinkingSettings},
    transport::MaxTokensParameter,
};

use super::{ProtocolModelClient, output_schema::apply_output_schema};

impl ProtocolModelClient {
    pub(super) fn build_wire_body(
        &self,
        messages: &[ModelMessage],
        settings: Option<&ModelSettings>,
        params: &ModelRequestParameters,
    ) -> Result<Value, ModelError> {
        let effective_settings = provider_policy_settings(settings, &self.profile);
        let settings = effective_settings.as_ref().or(settings);
        let mut body = match self.profile.protocol {
            ProtocolFamily::OpenAiChatCompletions => OpenAiChatAdapter::build_request_with_options(
                &self.model_name,
                messages,
                settings,
                &params.tools,
                self.openai_chat_max_tokens_parameter(),
            ),
            ProtocolFamily::OpenAiResponses => OpenAiResponsesAdapter::build_request_with_options(
                &self.model_name,
                messages,
                settings,
                &params.tools,
                &params.native_tools,
                self.openai_responses_max_tokens_parameter(),
            ),
            ProtocolFamily::AnthropicMessages => AnthropicMessagesAdapter::build_request(
                &self.model_name,
                messages,
                settings,
                &params.tools,
            ),
            ProtocolFamily::GeminiGenerateContent => {
                GeminiGenerateContentAdapter::build_request_with_native_tools(
                    messages,
                    settings,
                    &params.tools,
                    &params.native_tools,
                )
            }
            ProtocolFamily::BedrockConverse => BedrockConverseAdapter::build_request(
                &self.model_name,
                messages,
                settings,
                &params.tools,
            ),
        }?;
        apply_output_schema(&mut body, &self.profile, params);
        Ok(body)
    }

    fn openai_chat_max_tokens_parameter(&self) -> MaxTokensParameter {
        match self.http_config.max_tokens_parameter {
            MaxTokensParameter::Default if self.provider_name == "openai" => {
                MaxTokensParameter::MaxCompletionTokens
            }
            MaxTokensParameter::Default => MaxTokensParameter::MaxTokens,
            value => value,
        }
    }

    const fn openai_responses_max_tokens_parameter(&self) -> MaxTokensParameter {
        match self.http_config.max_tokens_parameter {
            MaxTokensParameter::Default => MaxTokensParameter::MaxOutputTokens,
            value => value,
        }
    }

    pub(super) fn parse_wire_response(&self, body: &Value) -> Result<ModelResponse, ModelError> {
        match self.profile.protocol {
            ProtocolFamily::OpenAiChatCompletions => OpenAiChatAdapter::parse_response(body),
            ProtocolFamily::OpenAiResponses => OpenAiResponsesAdapter::parse_response(body),
            ProtocolFamily::AnthropicMessages => AnthropicMessagesAdapter::parse_response(body),
            ProtocolFamily::GeminiGenerateContent => {
                GeminiGenerateContentAdapter::parse_response(body)
            }
            ProtocolFamily::BedrockConverse => BedrockConverseAdapter::parse_response(body),
        }
    }
}

fn provider_policy_settings(
    settings: Option<&ModelSettings>,
    profile: &ModelProfile,
) -> Option<ModelSettings> {
    let settings = settings?;
    if !profile.drop_sampling_parameters_when_reasoning
        || !settings.thinking.as_ref().is_some_and(thinking_is_active)
    {
        return None;
    }
    let mut policy_settings = settings.clone();
    policy_settings.temperature = None;
    policy_settings.top_p = None;
    policy_settings.top_k = None;
    policy_settings.presence_penalty = None;
    policy_settings.frequency_penalty = None;
    policy_settings.logit_bias.clear();
    Some(policy_settings)
}

fn thinking_is_active(thinking: &ThinkingSettings) -> bool {
    let mode_disabled = thinking
        .mode
        .as_deref()
        .is_some_and(|mode| matches!(mode, "disabled" | "off" | "none"));
    let effort_disabled = matches!(thinking.effort.as_str(), "" | "disabled" | "off" | "none");
    !mode_disabled && !effort_disabled
}