YouTip

Codex supports multiple models. Understand the characteristics of each model and choose the appropriate model based on your scenario. The desktop version can be switched in the bottom-right corner of the input box: !(#) Codex CLI can be switched using the /model command: !(#) * * * ## Available Models Codex currently supports the following models: | Model | Type | Characteristics | Applicable Scenarios | | --- | --- | --- | --- | | **gpt-5.4** | Flagship | Most powerful, deep reasoning | Complex tasks, architecture design | | **gpt-5.4-mini** | Lightweight | Fast response, lower cost | Simple tasks, rapid iteration | | **gpt-5.3-codex** | Professional | Programming optimization, code specialization | Code writing, bug fixes | | **gpt-5.3-codex-spark** | Fast | Lightning-fast response, high-frequency interaction | Real-time collaboration, quick Q&A | * * * ## Model Characteristics Detailed ### GPT-5.4 Flagship model with the strongest reasoning and creative capabilities. | Characteristic | Description | | --- | --- | | **Reasoning Depth** | In-depth analysis of complex problems | | **Context Understanding** | Understanding large codebase structures | | **Multi-step Tasks** | Handling complex workflows | | **Accuracy** | High accuracy rate, reducing rework | ### Applicable Scenarios * Architecture design and refactoring * Complex bug analysis and fixing * Multi-module coordinated development * Code review and quality analysis * * * ### GPT-5.4-mini Lightweight model with fast response, suitable for daily development. | Characteristic | Description | | --- | --- | | **Response Speed** | Faster than the flagship model | | **Cost Efficiency** | Lower token consumption | | **Daily Tasks** | Suitable for routine development operations | ### Applicable Scenarios * Simple feature implementation * Code formatting and refactoring * Documentation writing * Quick Q&A * * * ### GPT-5.3-Codex Model specially optimized for programming tasks. | Characteristic | Description | | --- | --- | | **Code Specialization** | Trained specifically for programming tasks | | **Language Coverage** | Supports multiple programming languages | | **Code Quality** | High-quality generated code | ### Applicable Scenarios * Code writing and generation * Bug fixing and debugging * Code refactoring * Test writing * * * ### GPT-5.3-Codex-Spark Lightning-fast response model, suitable for high-frequency interaction scenarios. | Characteristic | Description | | --- | --- | | **Lightning-fast Response** | Fastest response speed | | **Real-time Collaboration** | Suitable for interactive development | | **Pro Exclusive** | Only available in Pro plan | ### Applicable Scenarios * Real-time code Q&A * Rapid prototype validation * High-frequency iterative development * * * ## Reasoning Effort Configuration You can adjust the model's reasoning effort to balance speed and depth. ### Reasoning Effort Levels | Level | Description | Characteristics | | --- | --- | --- | | `minimal` | Minimal reasoning | Fastest response, suitable for simple tasks | | `low` | Low reasoning effort | Fast but with some analysis | | `medium` | Medium reasoning | Balanced speed and depth (default) | | `high` | High reasoning effort | In-depth analysis, suitable for complex tasks | | `xhigh` | Extra high reasoning | Strongest reasoning, slowest response | ### Configuring Reasoning Effort ## Reasoning Effort Settings # CLI Specification codex --reasoning-effort high # Configuration File model_reasoning_effort = "high" # Switch within session /model gpt-5.4 --reasoning-effort xhigh * * * ## Reasoning Summary Control the level of detail Codex displays for the reasoning process. ### Summary Modes | Mode | Description | | --- | --- | | `auto` | Automatically decide detail level (default) | | `concise` | Brief summary | | `detailed` | Detailed reasoning process | | `none` | No reasoning summary displayed | ## Reasoning Summary Settings # ~/.codex/config.toml model_reasoning_summary = "detailed" * * * ## Model Switching Switch between different models for different scenarios. ### Switching Methods ## Switching Models # CLI Slash Commands /model gpt-5.4 /model gpt-5.4-mini /model gpt-5.3-codex # With Reasoning Effort /model gpt-5.4--reasoning-effort high # In App Click the model selector and choose the target model ### Scenario Switching Suggestions | Scenario | Recommended Model | Reasoning Effort | | --- | --- | --- | | Architecture Design | gpt-5.4 | high/xhigh | | Complex Refactoring | gpt-5.4 | high | | Bug Fixing | gpt-5.3-codex | medium | | Daily Coding | gpt-5.4-mini | low/medium | | Quick Q&A | gpt-5.4-mini | minimal | | Real-time Collaboration | gpt-5.3-codex-spark | minimal | * * * ## Service Tiers Choosing different service tiers affects response priority. ### Service Tier Options | Tier | Description | | --- | --- | | `flex` | Flexible service, response may be slightly slower (default) | | `fast` | Priority service, faster response | ## Service Tier Settings # ~/.codex/config.toml service_tier = "fast" * * * ## Models and Plans Different plans have varying support for models: | Plan | Available Models | | --- | --- | | **Free** | Basic models | | **Plus** | gpt-5.4, gpt-5.4-mini, gpt-5.3-codex | | **Pro** | All models + Spark + Higher quotas | | **API Key** | Pay-per-token, supports mainstream models | * * * ## Cost Considerations ### Token Consumption Comparison | Model | Relative Cost | | --- | --- | | gpt-5.4 | Highest | | gpt-5.3-codex | Medium-high | | gpt-5.4-mini | Lower | | gpt-5.3-codex-spark | Low | ### Cost Optimization Suggestions * Use mini or Spark models for simple tasks * Use medium reasoning effort for daily development * Only use flagship models and high reasoning effort for complex tasks * Use /compact to compress context and reduce token consumption * * * ## Best Practices ### Model Selection Principles * Choose models based on task complexity * Balance response speed and reasoning depth * Prioritize accuracy for complex tasks * Prioritize response speed for high-frequency interactions ### Configuration Suggestions ## Recommended Configuration # Daily Development Configuration model = "gpt-5.4-mini" model_reasoning_effort = "medium" model_reasoning_summary = "auto" # Temporary Switch for Complex Tasks # /model gpt-5.4 --reasoning-effort high * * * ## FAQ ### Q: Which model is best for code writing? gpt-5.3-codex is specifically optimized for programming tasks and is suitable for most code writing scenarios. ### Q: How to balance speed and quality? Use gpt-5.4-mini + medium reasoning effort for daily tasks, and switch to gpt-5.4 + high reasoning effort for complex tasks. ### Q: What are the advantages of the Spark model? The Spark model has the fastest response time and is suitable for high-frequency interaction and real-time collaboration. It is only available in the Pro plan. ### Q: How does reasoning effort affect results? Higher reasoning effort means Codex will perform more in-depth analysis, but the response time will be longer.