Coding Models
AI models specialized for code generation and programming
Code
CodePrompt CacheReasoningTool useCached input 95% less
- Verified price-70% Discount
- Input$0.60/1MOutput$3.00/1M
- Official price
- Input$2.00/1MOutput$10.00/1M
CodeJSON modePrompt CacheReasoningTool useVisionCached input 90% less
- Verified price-70% Discount
- Input$0.03/1MOutput$0.15/1M
- Official price
- Input$0.10/1MOutput$0.50/1M
CodeJSON modePrompt CacheReasoningTool useVisionCached input 90% less
- Verified price-70% Discount
- Input$0.60/1MOutput$3.00/1M
- Official price
- Input$2.00/1MOutput$10.00/1M
CodeJSON modePrompt CacheReasoningTool useVisionCached input 90% less
- Verified price-70% Discount
- Input$0.06/1MOutput$0.36/1M
- Official price
- Input$0.20/1MOutput$1.20/1M
CodeJSON modePrompt CacheReasoningTool useVisionCached input 90% less
- Verified price-70% Discount
- Input$1.20/1MOutput$6.00/1M
- Official price
- Input$4.00/1MOutput$20.00/1M
CodeJSON modePrompt CacheReasoningTool useVisionCached input 90% less
- Verified price-70% Discount
- Input$0.60/1MOutput$3.60/1M
- Official price
- Input$2.00/1MOutput$12.00/1M
CodeJSON modePrompt CacheReasoningTool useVisionCached input 90% less
- Verified price-70% Discount
- Input$1.50/1MOutput$9.00/1M
- Official price
- Input$5.00/1MOutput$30.00/1M
CodeJSON modePrompt CacheReasoningTool useVisionCached input 90% less
- Verified price-70% Discount
- Input$3.00/1MOutput$15.00/1M
- Official price
- Input$10.00/1MOutput$50.00/1M
CodeJSON modePrompt CacheReasoningTool useSupports prompt caching
- Verified price— Discount
- —
- Official price
- Input$1.40/1MOutput$4.40/1M
CodeJSON modePrompt CacheReasoningTool useVisionSupports prompt caching
- Verified price— Discount
- —
- Official price
- Input$1.00/1MOutput$2.00/1M
CodeJSON modePrompt CacheReasoningTool useSupports prompt caching
- Verified price— Discount
- —
- Official price
- Input$0.4706/1MOutput$2.35/1M
CodeJSON modeReasoningTool useVision
- Verified price— Discount
- —
- Official price
- Input$30.00/1MOutput$180.00/1M
CodePrompt CacheReasoningTool useSupports prompt caching
- Verified price— Discount
- —
- Official price
- Input$3.00/1MOutput$15.00/1M
CodeReasoningTool useSupports prompt caching
- Verified price— Discount
- —
- Official price
- Input$0.30/1MOutput$2.50/1M
CodeReasoningTool useSupports prompt caching
- Verified price— Discount
- —
- Official price
- Input$1.50/1MOutput$7.50/1M
CodePrompt CacheSupports prompt caching
- Verified price— Discount
- —
- Official price
- Input$0.30/1MOutput$1.50/1M
All model series
6 seriesClaude 5
ChatVision+1
Auto from $0.03 · per 1M tokens
7 modelsOpen series
GPT-6
Featured seriesChatVision+1
Auto from $0.03 · per 1M tokens
4 modelsOpen series
Claude 4
Featured seriesChatVision+1
Auto from $0.30 · per 1M tokens
5 modelsOpen series
GPT 5.6
Featured seriesChatVision+1
Auto from $0.06 · per 1M tokens
3 modelsOpen series
GPT 5.4
Featured seriesChatVision+1
Auto from $0.14 · per 1M tokens
4 modelsOpen series
GPT 5.2
Featured seriesChatVision+1
Auto from $1.23 · per 1M tokens
2 modelsOpen series
Choose a coding model for the development loop
Evaluate a complete change, including reading the repository, editing, running tests, and recovering from errors. A fluent code snippet alone is not a completed development task.
Selection signals
- Check your client's request protocol, tool use, context limits, and supported output controls.
- Measure correct task completion, retries, latency, and total input/output cost on the same repository tasks.
- Keep test cases independent of the model's implementation and inspect the final diff before accepting changes.