Models

Claude model for creative writing, analysis, and controlled agent workflows

Multimodal Kimi model with 1M context and toggleable max-effort thinking for long-horizon agent work

Frontier GPT-5.6 model for complex professional work, coding, and agentic workflows

Everyday Claude agent model for coding, planning, browsing, and general work

Agent-ready GPT for coding and computer-use workflows at a lower cost

Balanced GPT-5.6 model for capable, cost-efficient everyday work

Cost-efficient GPT-5.6 model for fast, high-volume workloads

Default frontier GPT for coding, computer use, research, and knowledge work

Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work

Open MoE flagship with million-token context for coding and long agent runs

Fast Gemini model balancing multimodal reasoning, tool use, and cost

Original GPT-5 workhorse for reasoning, coding, writing, and tool workflows

Tencent Hy reasoning model for coding, instruction following, and agent tasks

Stronger MiMo Pro tier for multimodal reasoning and coding-agent execution

Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking

Multimodal Kimi workhorse for agent loops, coding tasks, and visual context

Newer StepFun flash model for faster agents, coding, and multimodal prompts

Earlier Kimi frontier model for long-context agents, coding, and multimodal work

Sharper GPT-5 generation for coding, product work, and tool-assisted tasks

Strong small GPT for coding subagents, quick tool use, and high-volume work

Hybrid-reasoning DeepSeek model with thinking and non-thinking modes, sparse attention, and tool-use

Cheapest GPT-5.4 lane for simple routing, extraction, and bulk automation

Open MiMo model for multimodal coding agents and long-context automation

Reasoning-first Gemini preview for agentic coding and complex problem solving

Google's proven reasoning model for coding, math, and multimodal analysis

Low-latency Gemini model for high-volume multimodal and agent workloads

Thinking Kimi model for slower research passes, planning, and hard technical questions