Struct CustomProviderConfig 

pub struct CustomProviderConfig {

    pub name: String,
    pub display_name: String,
    pub base_url: String,
    pub api_format: CustomProviderApiFormat,
    pub context_window: Option<usize>,
    pub supports_tools: Option<bool>,
    pub supports_reasoning: Option<bool>,
    pub supports_reasoning_effort: Option<bool>,
    pub supports_vision: Option<bool>,
    pub supports_structured_output: Option<bool>,
    pub supports_parallel_tool_calls: Option<bool>,
    pub supports_context_caching: Option<bool>,
    pub supports_responses_compaction: Option<bool>,
    pub supports_context_edits: Option<bool>,
    pub temperature: Option<f32>,
    pub top_p: Option<f32>,
    pub top_k: Option<i32>,
    pub presence_penalty: Option<f32>,
    pub frequency_penalty: Option<f32>,
    pub max_tokens: Option<u32>,
    pub reasoning_effort: Option<ReasoningEffortLevel>,
    pub api_key_env: String,
    pub auth: Option<CustomProviderCommandAuthConfig>,
    pub model: String,
    pub models: Vec<String>,
    pub profiles: BTreeMap<String, CustomProviderProfileConfig>,
}

Configuration for a user-defined OpenAI-compatible provider endpoint.

Allows users to define multiple named custom endpoints (e.g., corporate proxies) with distinct display names, so they can toggle between them and clearly see which endpoint is active.

Fields

name: String

Stable provider key used for routing and persistence (e.g., “mycorp”). Must be lowercase alphanumeric with optional hyphens/underscores.

display_name: String

Human-friendly label shown in the TUI header, footer, and model picker (e.g., “MyCorporateName”).

base_url: String

Base URL of the OpenAI-compatible API endpoint (e.g., <https://llm.corp.example/v1>). Non-empty custom providers from repository-controlled workspace/project layers are rejected.

api_format: CustomProviderApiFormat

Typed API format for the provider’s default profile.

context_window: Option<usize>

Optional context window size in tokens for models served by this endpoint.

When omitted, the OpenAI-compatible provider uses its default context window size.

supports_tools: Option<bool>

Optional support for tool calling.

supports_reasoning: Option<bool>

Optional support for reasoning.

supports_reasoning_effort: Option<bool>

Optional support for reasoning effort.

supports_vision: Option<bool>

Optional support for vision inputs.

supports_structured_output: Option<bool>

Optional support for structured output.

supports_parallel_tool_calls: Option<bool>

Optional support for parallel tool calls.

supports_context_caching: Option<bool>

Optional support for context caching.

supports_responses_compaction: Option<bool>

Optional support for responses compaction.

supports_context_edits: Option<bool>

Optional support for context edits.

temperature: Option<f32>

Optional sampling temperature default (0.0-2.0) for models served by this endpoint unless a profile overrides it.

top_p: Option<f32>

Optional nucleus-sampling default (0.0-1.0).

top_k: Option<i32>

Optional top-k default (>= 0).

presence_penalty: Option<f32>

Optional presence penalty default (-2.0-2.0).

frequency_penalty: Option<f32>

Optional frequency penalty default (-2.0-2.0).

max_tokens: Option<u32>

Optional max output tokens default (> 0).

reasoning_effort: Option<ReasoningEffortLevel>

Optional reasoning effort default for models without a profile override.

api_key_env: String

Environment variable name that holds the API key for this endpoint (e.g., “MYCORP_API_KEY”).

auth: Option<CustomProviderCommandAuthConfig>

Optional command-backed bearer token configuration.

model: String

Default model to use with this endpoint (e.g., “gpt-5-mini”).

When models is empty, this single model is what the /model picker offers for this provider. When models is non-empty, this field is used as the default selection but the picker lists every entry in models.

models: Vec<String>

Optional list of additional model identifiers offered by the provider.

Useful for OpenAI-compatible aggregators such as Atlas Cloud that expose many models behind a single endpoint. When set, the /model picker shows one entry per model. When empty, the picker falls back to the single model field.

profiles: BTreeMap<String, CustomProviderProfileConfig>

Exact model-keyed sparse capability profiles.

Implementations

impl CustomProviderConfig

pub fn resolved_api_key_env(&self) -> String

Resolve the API key environment variable used for this provider.

Falls back to a derived NAME_API_KEY-style variable when the config does not set api_key_env.

pub fn uses_command_auth(&self) -> bool

pub fn effective_models(&self) -> Vec<String>

Return the list of models the /model picker should offer for this provider.

If models is non-empty, every entry is returned (trimmed). Otherwise the single model field is returned as a one-element list. An empty model field with no models list yields an empty result.

pub fn profile(&self, model: &str) -> Option<&CustomProviderProfileConfig>

pub fn resolved_profile(&self, model: &str) -> ResolvedCustomProviderProfile

pub fn provider_defaults_profile(&self) -> CustomProviderProfileConfig

pub fn validate(&self) -> Result<(), String>

Validate that required fields are present and the name doesn’t collide with built-in provider keys.

Trait Implementations

impl Clone for CustomProviderConfig

fn clone(&self) -> Self

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for CustomProviderConfig

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl Default for CustomProviderConfig

fn default() -> Self

Returns the “default value” for a type.

impl<'de> Deserialize<'de> for CustomProviderConfig

fn deserialize<__D>(__deserializer: __D) -> Result<Self, __D::Error>

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl JsonSchema for CustomProviderConfig

fn schema_name() -> Cow<'static, str>

The name of the generated JSON Schema.

fn schema_id() -> Cow<'static, str>

Returns a string that uniquely identifies the schema produced by this type.

fn json_schema(generator: &mut SchemaGenerator) -> Schema

Generates a JSON Schema for this type.

fn inline_schema() -> bool

Whether JSON Schemas generated for this type should be included directly in parent schemas, rather than being re-used where possible using the $ref keyword.

impl PartialEq for CustomProviderConfig

fn eq(&self, other: &Self) -> bool

Equality operator ==.

fn ne(&self, other: &Rhs) -> bool

Inequality operator !=.

impl Serialize for CustomProviderConfig

fn serialize<__S>(&self, __serializer: __S) -> Result<__S::Ok, __S::Error>

where __S: Serializer,

Serialize this value into the given Serde serializer.

impl StructuralPartialEq for CustomProviderConfig