89 AI Models
One Platform

Anthropic's newest Claude 5 Sonnet model. Near-Opus quality on coding and agentic work at Sonnet cost, with adaptive extended thinking and a 1M-token context window.

Anthropic's Claude 4.6 Sonnet model. Delivers enhanced performance across coding, analysis, and complex reasoning tasks with improved speed and efficiency.

Anthropic's most intelligent model in the Claude 4 family.

Anthropic's latest flagship model. Claude 5.5 Opus delivers adaptive extended thinking, coding, and agentic capabilities at a lower price than Claude 5 Opus.

Anthropic's previous flagship model. Claude 5 Opus approaches Fable 5 performance at Opus 4.8 pricing, with adaptive extended thinking, coding, and agentic capabilities.

Anthropic's previous flagship model. Claude 4.8 Opus delivers enhanced frontier intelligence with improved extended thinking, coding, and agentic capabilities over 4.7.

Anthropic's previous flagship model. Claude 4.7 Opus delivers frontier intelligence with extended thinking, coding, and agentic capabilities.

Anthropic's earlier flagship model. Claude 4.6 Opus delivers frontier intelligence with extended thinking, coding, and agentic capabilities.

Claude 4.5 Opus with Anthropic. Top-tier extended thinking model with excellent performance for complex reasoning, coding, and creative tasks.

Anthropic's fastest model. Claude 5.5 Haiku targets high-volume, latency-sensitive work such as classification, extraction, routing and subagents, with adaptive thinking and a 1M-token context window.

Anthropic's Claude Fable 5.1 - the current Fable release, built for complex, long-running asynchronous and agentic work.

Anthropic's most capable model, built for complex, long-running asynchronous tasks.

OpenAI's GPT-6 Astra - the frontier of the GPT-6 family.

GPT-5.6 Terra - the balanced GPT-5.6 variant. Strong reasoning and vision, priced between Sol and Luna, a reliable workhorse for everyday professional tasks.

OpenAI's GPT-5.6 Sol - flagship of the 5.6 family.

GPT-5.6 Luna - the fast, cost-efficient GPT-5.6 variant.

OpenAI's next-generation flagship GPT-5.5 model. A new class of intelligence for coding and professional work with a 1M+ context window, advanced reasoning, vision, and tool use.

Ultra-lightweight GPT-5.4 model optimized for speed and cost efficiency.

Compact GPT-5.4 variant balancing strong performance with lower cost.

High-performing GPT-5.4 model with strong reasoning, creativity, and vision understanding.

Previous-generation GPT-5 flagship with robust multimodal understanding and deep analytical capabilities.

Early GPT-5 series model with solid reasoning and vision support.

Ultra-efficient GPT-5 variant designed for lightweight tasks and edge deployment.

Efficient multimodal model with vision and tool support, suitable for cost-sensitive tasks requiring image analysis and extended context.

OpenAI's advanced language model with exceptional reasoning, creativity, and multimodal capabilities.

OpenAI's cost-effective multimodal model with vision capabilities.

OpenAI's multimodal model with strong vision capabilities and reasoning.

OpenAI GPT-Image-2.5 Sunburst - OpenAI's most capable image generation and editing model, best where editing precision matters most.

OpenAI GPT-Image-2.5 Flare - Fast, high-quality everyday image generation and editing.

OpenAI GPT-Image-2 - State-of-the-art image generation with flexible resolutions up to 4K, improved quality, and fast generation.

OpenAI GPT-Image-1.5 - Enhanced version of GPT-Image-1 with improved image generation quality and text integration, supporting up to 2048x2048 resolution and image editing capabilities.

OpenAI GPT-Image-1 Mini - Faster, cost-effective version of GPT-Image-1 with good quality image generation and text integration, supporting up to 2048x2048 resolution and image editing capabilities.

Google's most intelligent Flash model, delivering sustained frontier performance on agentic and coding tasks, built on state-of-the-art reasoning.

Google's Gemini 3.1 Pro preview for multimodal understanding, delivering richer visuals and deeper interactivity, built on a foundation of state-of-the-art reasoning.

Google's most cost-efficient Gemini model, optimized for low-latency, high-volume tasks while matching Gemini 2.5 Flash quality across key capabilities.

Google's Gemini 3 Flash preview for fast, low-latency multimodal understanding, delivering richer visuals and deeper interactivity, built on a foundation of state-of-the-art reasoning.

Google's state-of-the-art image generation and editing model (Nano Banana Pro).

Google's high-efficiency image generation and editing model (Nano Banana 2).

Google's Gemini 2.5 Flash image generation and editing model (Nano Banana).

xAI's Grok 4.5 reasoning model. 500K context window with vision, tool use, and extended reasoning.

Moonshot's Kimi K3 flagship. 1M context with native vision, tool use, and selectable reasoning effort (low/high/max).

Kimi K2.7 Code served at 180-260 tokens/s for latency-sensitive work.

Moonshot's coding-focused Kimi, tuned for long-horizon repository work with less overthinking.

Moonshot's multimodal workhorse for agent loops, coding, and visual context.

DeepSeek's V4-Pro, the heavier reasoning tier of the V4 line.

DeepSeek's V4.1-Flash. 1M context with native vision, tool use, and selectable reasoning effort (low/high/max).

FLUX Pro Ultra - Premium image generation with enhanced capabilities and exceptional quality output.

FLUX Pro 1.1 - Professional quality image generation with strong prompt adherence and fast turnaround.

FLUX Kontext Pro - Advanced image-to-image transformation with high-quality output and versatile editing capabilities.

FLUX Kontext Max - Premium image-to-image transformation with maximum quality and capabilities.

Meta's powerful open-source Llama 3 model with 70B parameters.

Meta's open-source Llama 3 model with 8B parameters.

Moonshot's Kimi K2.5 via AWS Bedrock - multimodal, 256K context, no provider key required.

Moonshot's Kimi K2 Thinking via AWS Bedrock - chain-of-thought reasoning for math, coding and logic.

DeepSeek's latest model via AWS Bedrock. Large context window, strong reasoning, and high throughput.

Claude 4.1 Opus via AWS Bedrock. Latest iteration with improved performance and reliability.

Claude 4 Opus via AWS Bedrock. Anthropic's most capable model with enhanced reasoning and multimodal capabilities.