AI Models
Choose from our range of AI models optimized for different tasks. All models are available across all subscription tiers.
Understanding Compute Tokens (CT)
Compute Tokens (CT) are our unified billing unit for both AI model usage and environment/container runtime. LLM API costs are passed through at a 1:1 rate with zero markup—we don't add margins on top of provider pricing. The CT Cost column shows relative cost compared to Claude Sonnet 4.5 (our 1x baseline).
Model Comparison
| Model | Intelligence | Speed | Context | Input / 1M | Cached / 1M | Output / 1M | CT Cost |
|---|---|---|---|---|---|---|---|
Claude Opus 4.7 Anthropic’s latest flagship for the most complex reasoning and agentic coding work. | Highest | Medium | 1M | $5.00 | $0.50 | $25.00 | 5.0x |
Claude Opus 4.6 Previous-generation flagship Claude model for complex reasoning and coding. | Highest | Medium | 1M | $5.00 | $0.50 | $25.00 | 5.0x |
Claude Sonnet 4.5 Best balance of intelligence and speed. Excellent for coding and daily tasks. | High | Fast | 200k | $3.00 | $0.30 | $15.00 | 3.0x |
Claude Haiku 4.5Recommended Fast and cost-effective for quick tasks and high-volume operations. | Good | Very Fast | 200k | $1.00 | $0.10 | $5.00 | 1x (baseline) |
GPT-5.5 Pro OpenAI highest-accuracy model for the hardest professional and agentic work. | Highest | Medium | 1M | $30.00 | — | $180.00 | 35.1x |
GPT-5.5 OpenAI frontier model for coding, professional work, and long-context agents. | Highest | Fast | 1M | $5.00 | $0.50 | $30.00 | 5.6x |
GPT-5.4 OpenAI flagship model for advanced coding, planning, and computer use. | Highest | Fast | 1M | $2.50 | $0.25 | $15.00 | 2.8x |
GPT-5.4 mini OpenAI mini model tuned for fast coding, subagents, and computer-use tasks. | High | Fast | 400k | $0.75 | $0.07 | $4.50 | 0.84x |
GPT-5.4 nano Lightweight OpenAI model for simple instructions, routing, and high-volume tasks. | Good | Very Fast | 400k | $0.20 | $0.02 | $1.25 | 0.23x |
Gemini 3.1 Pro Higher-reasoning Gemini model for complex planning and analysis. | Highest | Fast | 1M | $3.00 | $0.30 | $15.00 | 3.0x |
Gemini 3 Flash Fast default model for everyday agent execution. | High | Very Fast | 1M | $1.00 | $0.10 | $5.00 | 1x (baseline) |
Gemini 3.1 Flash Fast Gemini tier that currently routes through the same runtime as Gemini 3 Flash. | High | Very Fast | 1M | $1.00 | $0.10 | $5.00 | 1x (baseline) |
DeepSeek V4 Pro DeepSeek flagship model running through Clawcode for agentic coding and long-context reasoning. | Highest | Fast | 1M | $0.43 | $0.04 | $0.87 | 0.28x |
DeepSeek V4 Flash Fast DeepSeek V4 model running through Clawcode for efficient agent execution. | High | Very Fast | 1M | $0.14 | $0.03 | $0.28 | 0.09x |
Kimi K2.6 Moonshot flagship served via Cloudflare Workers AI for long-horizon coding and agentic execution. | Highest | Fast | 262k | $0.95 | $0.16 | $4.00 | 0.87x |
Research Models
Specialized models for comprehensive web research, multi-source analysis, and detailed report generation.
| Model | Intelligence | Speed | Context | Input / 1M | Output / 1M | Best For |
|---|---|---|---|---|---|---|
Gemini 3 ProRecommended Most capable research model with Deep Think reasoning and 1M token context. | Highest | Medium | 1M | $2.00 | $12.00 | Deep researchComplex reasoningComprehensive reports |
Gemini 3 Flash Fast research model, 3x faster than Pro with excellent cost efficiency. | Highest | Fast | 1M | $0.50 | $3.00 | Quick researchSummariesHigh-volume tasks |
Image Generation Models
Create stunning visuals with our AI image generation models, from rapid prototypes to photorealistic artwork.
| Model | Quality | Speed | Max Resolution | Price / Image | Best For |
|---|---|---|---|---|---|
Gemini 3 Pro ImageRecommended Premium image generation with up to 4K resolution and exceptional detail. | Highest | Medium | Up to 4096x4096 | $0.134 2K $0.24 4K | Professional graphicsMarketing assetsHigh-res artwork |
Gemini 3 Flash Image Fast image generation for rapid prototyping and high-volume workflows. | High | Very Fast | Up to 1024x1024 | $0.039 | Quick mockupsSocial mediaBatch generation |
Flagship Models
Our most capable models with the highest intelligence and reasoning capabilities.
Claude Opus 4.7
Anthropic’s latest flagship for the most complex reasoning and agentic coding work.
GPT-5.5 Pro
OpenAI highest-accuracy model for the hardest professional and agentic work.
GPT-5.5
OpenAI frontier model for coding, professional work, and long-context agents.
GPT-5.4
OpenAI flagship model for advanced coding, planning, and computer use.
Gemini 3.1 Pro
Higher-reasoning Gemini model for complex planning and analysis.
Kimi K2.6
Moonshot flagship served via Cloudflare Workers AI for long-horizon coding and agentic execution.
DeepSeek V4 Pro
DeepSeek flagship model running through Clawcode for agentic coding and long-context reasoning.
Claude Opus 4.6
Previous-generation flagship Claude model for complex reasoning and coding.
Claude Sonnet 4.5
Best balance of intelligence and speed. Excellent for coding and daily tasks.
Efficient Models
Fast and cost-effective for high-volume operations and simpler tasks.
GPT-5.4 mini
OpenAI mini model tuned for fast coding, subagents, and computer-use tasks.
GPT-5.4 nano
Lightweight OpenAI model for simple instructions, routing, and high-volume tasks.
Gemini 3 Flash
Fast default model for everyday agent execution.
Gemini 3.1 Flash
Fast Gemini tier that currently routes through the same runtime as Gemini 3 Flash.
DeepSeek V4 Flash
Fast DeepSeek V4 model running through Clawcode for efficient agent execution.
Claude Haiku 4.5
RecommendedFast and cost-effective for quick tasks and high-volume operations.
About Pricing
All prices are in USD per 1 million tokens. Input tokens are charged when sending prompts to the model. Cached tokens are charged when using prompt caching (if supported). Output tokens are charged for model responses. Your actual cost depends on the model used and the complexity of your tasks.View subscription plans