Enum ChatModel 

#[non_exhaustive]
pub enum ChatModel {
    Gpt5_6,
    Gpt5_6Sol,
    Gpt5_6Terra,
    Gpt5_6Luna,
    Gpt5_5,
    Gpt5_5Pro,
    Gpt5_4,
    Gpt5_4Pro,
    Gpt5_4Mini,
    Gpt5_4Nano,
    Gpt5_3ChatLatest,
    Gpt5_3Codex,
    Gpt5_2Codex,
    Gpt5_1Codex,
    Gpt5SearchApi,
    Gpt4oSearchPreview,
    Gpt4oMiniSearchPreview,
    GptAudio,
    GptAudio1_5,
    GptAudioMini,
    Gpt5,
    Gpt5Pro,
    Gpt5_2,
    Gpt5_2ChatLatest,
    Gpt5_2Pro,
    Gpt5_1,
    Gpt5_1ChatLatest,
    Gpt5_1CodexMax,
    Gpt5Mini,
    Gpt5Nano,
    Gpt4_1,
    Gpt4_1Mini,
    Gpt4_1Nano,
    Gpt4o,
    Gpt4oMini,
    Gpt4oAudioPreview,
    Gpt4Turbo,
    Gpt4,
    Gpt3_5Turbo,
    Gpt3_5Turbo16k,
    O1,
    O1Pro,
    O3,
    O3Pro,
    O3Mini,
    O4Mini,
    Custom(String),
}

Models available for Chat Completions and Responses APIs.

This enum covers all models that can be used with the Chat Completions API (/v1/chat/completions) and the Responses API (/v1/responses).

Model Categories

GPT-5 Series (Latest Flagship)

• [Gpt5_2]: GPT-5.2 Thinking - flagship model for coding and agentic tasks
• [Gpt5_2ChatLatest]: GPT-5.2 Instant - fast workhorse for everyday work
• [Gpt5_2Pro]: GPT-5.2 Pro - smartest for difficult questions (Responses API only)
• [Gpt5_1]: GPT-5.1 - configurable reasoning and non-reasoning
• [Gpt5_1CodexMax]: GPT-5.1 Codex Max - powers Codex CLI
• [Gpt5Mini]: GPT-5 Mini - smaller, faster variant

GPT-4.1 Series

• [Gpt4_1]: 1M context window flagship
• [Gpt4_1Mini]: Balanced performance and cost
• [Gpt4_1Nano]: Fastest and most cost-efficient

GPT-4o Series

• [Gpt4o]: High-intelligence flagship model
• [Gpt4oMini]: Cost-effective GPT-4o variant
• [Gpt4oAudioPreview]: Audio-capable GPT-4o

Reasoning Models (o-series)

• [O1], [O1Pro]: Full reasoning models
• [O3], [O3Mini]: Latest reasoning models
• [O4Mini]: Fast, cost-efficient reasoning

Reasoning Model Restrictions

Reasoning models (GPT-5 series, o1, o3, o4 series) have parameter restrictions:

• temperature: Only 1.0 supported
• top_p: Only 1.0 supported
• frequency_penalty: Only 0 supported
• presence_penalty: Only 0 supported

GPT-5 models support reasoning.effort parameter:

• none: No reasoning (GPT-5.1 default)
• minimal: Very few reasoning tokens
• low, medium, high: Increasing reasoning depth
• xhigh: Maximum reasoning (GPT-5.2 Pro, GPT-5.1 Codex Max)

Example

use openai_tools::common::models::ChatModel;

// Check if a model is a reasoning model
let model = ChatModel::O3Mini;
assert!(model.is_reasoning_model());

// GPT-5 models are also reasoning models
let gpt5 = ChatModel::Gpt5_2;
assert!(gpt5.is_reasoning_model());

// Get the API model ID string
assert_eq!(model.as_str(), "o3-mini");

Variants (Non-exhaustive)

Non-exhaustive enums could have additional variants added in future. Therefore, when matching against variants of non-exhaustive enums, an extra wildcard arm must be added to account for any future variants.

Gpt5_6

GPT-5.6 - Alias routing to GPT-5.6 Sol

Gpt5_6Sol

GPT-5.6 Sol - Frontier model for complex professional work

• Context: 1.05M tokens, 128K max output
• Knowledge cutoff: 2026-02-16
• Supports: reasoning.effort

Gpt5_6Terra

GPT-5.6 Terra - Balanced model for everyday work

• Context: 1.05M tokens, 128K max output
• Supports: reasoning.effort

Gpt5_6Luna

GPT-5.6 Luna - Most cost-efficient GPT-5.6 variant

• Context: 1.05M tokens, 128K max output
• Supports: reasoning.effort

Gpt5_5

GPT-5.5 - Frontier model with 1M context

• Context: 1.05M tokens, 128K max output
• Supports: reasoning.effort (none, low, medium (default), high, xhigh)

Gpt5_5Pro

GPT-5.5 Pro - Uses more compute for consistently better answers

• Available in Responses API only (no Chat Completions)
• Supports: reasoning.effort (medium, high (default), xhigh)

Gpt5_4

GPT-5.4 - Frontier model for complex professional work

• Context: 1.05M tokens
• Supports: reasoning.effort (none (default), low, medium, high, xhigh)

Gpt5_4Pro

GPT-5.4 Pro - Uses more compute for consistently better answers

• Available in Responses API only (no Chat Completions)
• Supports: reasoning.effort (medium (default), high, xhigh)

Gpt5_4Mini

GPT-5.4 Mini - Strong mini model for coding, computer use and subagents

• Context: 400K tokens
• Supports: reasoning.effort

Gpt5_4Nano

GPT-5.4 Nano - Cheapest GPT-5.4-class model for high-volume tasks

• Context: 400K tokens
• Supports: reasoning.effort (none (default), low, medium, high, xhigh)

Gpt5_3ChatLatest

GPT-5.3 Instant - Non-reasoning chat model

Points to the GPT-5.3 Instant snapshot used in ChatGPT.

Deprecated by OpenAI - GPT-5.6 is recommended for most API usage.

Gpt5_3Codex

GPT-5.3 Codex - Most capable agentic coding model

• Available in Responses API only (no Chat Completions)
• Context: 400K tokens, 128K max output
• Supports: reasoning.effort (low, medium, high, xhigh)

Gpt5_2Codex

GPT-5.2 Codex - GPT-5.2 optimized for agentic coding tasks

• Available in Responses API only (no Chat Completions)
• Context: 400K tokens
• Supports: reasoning.effort (low, medium, high, xhigh)

Gpt5_1Codex

GPT-5.1 Codex - GPT-5 optimized for agentic coding tasks

• Available in Responses API only (no Chat Completions)

Gpt5SearchApi

GPT-5 Search API - GPT-5 with built-in web search

Rejects sampling parameters and exposes no reasoning parameter.

Gpt4oSearchPreview

GPT-4o Search Preview - GPT-4o with built-in web search

Gpt4oMiniSearchPreview

GPT-4o Mini Search Preview - GPT-4o Mini with built-in web search

GptAudio

GPT Audio - audio-capable chat model

Requires an audio input content part or an audio output modality.

GptAudio1_5

GPT Audio 1.5 - newer audio-capable chat model

GptAudioMini

GPT Audio Mini - cost-efficient audio-capable chat model

Gpt5

GPT-5 - Original GPT-5 flagship reasoning model

Gpt5Pro

GPT-5 Pro - GPT-5 with more compute for difficult questions

• Available in Responses API only (no Chat Completions)

Gpt5_2

GPT-5.2 Thinking - Flagship model for coding and agentic tasks

• Context: 128K tokens (256K with thinking)
• Supports: reasoning.effort (none, minimal, low, medium, high, xhigh)
• Supports: verbosity parameter (low, medium, high)

Gpt5_2ChatLatest

GPT-5.2 Instant - Fast workhorse for everyday work

Points to the GPT-5.2 Instant snapshot used in ChatGPT. This is a non-reasoning model, so it accepts the full standard parameter set.

Deprecated by OpenAI - GPT-5.6 is recommended for most API usage.

Gpt5_2Pro

GPT-5.2 Pro - Smartest for difficult questions

• Available in Responses API only
• Supports: xhigh reasoning effort

Gpt5_1

GPT-5.1 - Configurable reasoning and non-reasoning

• Defaults to no reasoning (effort: none)
• Supports: reasoning.effort (none, low, medium, high)

Gpt5_1ChatLatest

GPT-5.1 Instant - Chat-optimized GPT-5.1

Points to the GPT-5.1 Instant snapshot used in ChatGPT. This is a non-reasoning model, so it accepts the full standard parameter set.

Gpt5_1CodexMax

GPT-5.1 Codex Max - Powers Codex and Codex CLI

• Available in Responses API only
• Supports: reasoning.effort (none, medium, high, xhigh)

Gpt5Mini

GPT-5 Mini - Smaller, faster GPT-5 variant

Gpt5Nano

GPT-5 Nano - Fastest, most cost-efficient GPT-5 variant

Gpt4_1

GPT-4.1 - Smartest non-reasoning model with 1M token context

Gpt4_1Mini

GPT-4.1 Mini - Balanced performance and cost

Gpt4_1Nano

GPT-4.1 Nano - Fastest and most cost-efficient

Gpt4o

GPT-4o - High-intelligence flagship model (multimodal)

Gpt4oMini

GPT-4o Mini - Cost-effective GPT-4o variant

Gpt4oAudioPreview

GPT-4o Audio Preview - Audio-capable GPT-4o

Gpt4Turbo

GPT-4 Turbo - High capability with faster responses

Gpt4

GPT-4 - Original GPT-4 model

Gpt3_5Turbo

GPT-3.5 Turbo - Fast and cost-effective

Gpt3_5Turbo16k

GPT-3.5 Turbo 16K - legacy extended-context GPT-3.5 Turbo

O1

O1 - Full reasoning model for complex tasks

O1Pro

O1 Pro - O1 with more compute for complex problems

O3

O3 - Latest full reasoning model

O3Pro

O3 Pro - O3 with more compute

• Available in Responses API only (no Chat Completions)

O3Mini

O3 Mini - Smaller, faster reasoning model

O4Mini

O4 Mini - Fast, cost-efficient reasoning model

Custom(String)

Custom model ID for fine-tuned models or new models not yet in enum

Implementations

impl ChatModel

pub fn as_str(&self) -> &str

Returns the model identifier string for API requests.

Example

use openai_tools::common::models::ChatModel;

assert_eq!(ChatModel::Gpt4oMini.as_str(), "gpt-4o-mini");
assert_eq!(ChatModel::O3Mini.as_str(), "o3-mini");
assert_eq!(ChatModel::Gpt5_2.as_str(), "gpt-5.2");

pub fn is_reasoning_model(&self) -> bool

Checks if this is a reasoning model with parameter restrictions.

Reasoning models (GPT-5 series, o1, o3, o4 series) only support:

• temperature = 1.0
• top_p = 1.0
• frequency_penalty = 0
• presence_penalty = 0

Example

use openai_tools::common::models::ChatModel;

assert!(ChatModel::O3Mini.is_reasoning_model());
assert!(ChatModel::Gpt5_2.is_reasoning_model());
assert!(!ChatModel::Gpt4oMini.is_reasoning_model());
assert!(!ChatModel::Gpt4_1.is_reasoning_model());

pub fn is_search_model(&self) -> bool

Checks if this is a web-search model.

Search models (gpt-5-search-api, gpt-4o-search-preview, gpt-4o-mini-search-preview) reject the whole sampling parameter set - temperature, top_p, n, logprobs and the penalties - but unlike reasoning models they expose no reasoning parameter.

Example

use openai_tools::common::models::ChatModel;

assert!(ChatModel::Gpt4oSearchPreview.is_search_model());
assert!(!ChatModel::Gpt4oSearchPreview.is_reasoning_model());
assert!(!ChatModel::Gpt4oMini.is_search_model());

pub fn parameter_support(&self) -> ParameterSupport

Returns parameter support information for this model.

This method provides detailed information about which parameters are supported by the model and any restrictions that apply.

Example

use openai_tools::common::models::{ChatModel, ParameterRestriction};

// Standard model supports all parameters
let standard = ChatModel::Gpt4oMini;
let support = standard.parameter_support();
assert_eq!(support.temperature, ParameterRestriction::Any);
assert!(support.logprobs);

// Reasoning model has restrictions
let reasoning = ChatModel::O3Mini;
let support = reasoning.parameter_support();
assert_eq!(support.temperature, ParameterRestriction::FixedValue(1.0));
assert!(!support.logprobs);
assert!(support.reasoning);

pub fn custom(model_id: impl Into<String>) -> Self

Creates a custom model from a string.

Use this for fine-tuned models or new models not yet in the enum.

Example

use openai_tools::common::models::ChatModel;

let model = ChatModel::custom("ft:gpt-4o-mini:my-org::abc123");
assert_eq!(model.as_str(), "ft:gpt-4o-mini:my-org::abc123");

Trait Implementations

impl Clone for ChatModel

fn clone(&self) -> ChatModel

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for ChatModel

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

Formats the value using the given formatter.

impl Default for ChatModel

fn default() -> ChatModel

Returns the “default value” for a type.

impl<'de> Deserialize<'de> for ChatModel

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

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl Display for ChatModel

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

Formats the value using the given formatter.

impl Eq for ChatModel

impl From<&str> for ChatModel

fn from(s: &str) -> Self

Converts to this type from the input type.

impl From<String> for ChatModel

fn from(s: String) -> Self

Converts to this type from the input type.

impl Hash for ChatModel

fn hash<__H: Hasher>(&self, state: &mut __H)

Feeds this value into the given Hasher.

fn hash_slice<H>(data: &[Self], state: &mut H)

where H: Hasher, Self: Sized,

Feeds a slice of this type into the given Hasher.

impl PartialEq for ChatModel

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

Equality operator ==.

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

Inequality operator !=.

impl Serialize for ChatModel

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 ChatModel