Models & pricing
Browse every model.
Compare published input and output rates, context limits, and capabilities. Point your OpenAI-compatible client at https://api.unbar.sh/v1 and choose any model per request.
Alibaba Qwen vision-language model with efficient hybrid attention and expert routing.
Alibaba Qwen model for cost-efficient multimodal coding and reasoning.
Anthropic Mythos-class model for autonomous knowledge work and coding.
Anthropic frontier model for agentic coding, long-running workflows, and knowledge work.
Anthropic fast model for efficient general-purpose work.
Anthropic fast model for high-volume work such as summarization and subagents.
Anthropic Opus model for coding and long-running professional tasks.
Anthropic Opus model for long-running asynchronous agents and coding.
Anthropic general-purpose multimodal model for long-running agents and software work.
Anthropic flagship for demanding reasoning, coding, and long-horizon agentic work.
Anthropic flagship for demanding reasoning, large-codebase changes, and long-horizon agentic work.
Anthropic Sonnet model for coding, agents, and professional work.
Anthropic Sonnet model for coding, agents, and adaptive reasoning.
Anthropic Sonnet model for building features, fixing bugs, and everyday agentic work.
DeepSeek efficiency-optimized mixture-of-experts model for fast inference and coding.
DeepSeek large-scale mixture-of-experts model for advanced reasoning and coding.
Google workhorse model for advanced reasoning, coding, and scientific tasks.
Google advanced reasoning, coding, mathematics, and scientific model.
Google frontier reasoning model for software engineering and complex workflows.
Google efficient multimodal model for coding and parallel agentic execution.
Google efficient Flash model for focused multi-agent workflows.
Google high-efficiency model for coding and agentic workflows.
Google multimodal Flash model for fast agentic workflows and coding.
Google most intelligent Flash model for software engineering and agentic tasks.
MiniMax compact model for end-to-end coding and agentic workflows.
MiniMax productivity model for complex real-world digital work.
MiniMax multimodal foundation model for long-horizon agentic work and coding.
Moonshot mixture-of-experts language model optimized for coding and reasoning.
Moonshot native multimodal model for visual coding and agent swarms.
Moonshot multimodal reasoning model for coding and long-horizon agentic workflows.
OpenAI multimodal model for text and image inputs with text output.
OpenAI compact multimodal model for efficient general-purpose work.
OpenAI frontier model for complex professional workloads and hard reasoning.
OpenAI cost-efficient model for high-volume, latency-sensitive tasks.
OpenAI flagship for complex reasoning, coding, and agentic workflows.
OpenAI balanced model for everyday coding, reasoning, and agentic work.
OpenAI flagship for advanced analysis, software engineering, and deep research.
OpenAI fast, cost-efficient model for high-volume, latency-sensitive tasks.
OpenAI cost-efficient high-end model for demanding professional workloads.
OpenAI reasoning model for math, science, coding, and visual reasoning.
OpenAI reasoning model with additional compute for difficult answers.
OpenAI compact reasoning model for fast, cost-efficient multimodal work.
xAI reasoning model for high-speed agentic tool calling.
xAI reasoning model for agentic workflows and factual tasks.
xAI frontier model for coding, knowledge work, and STEM.
xAI frontier model for coding, knowledge work, and STEM.
Z.ai flagship model for programming and stable multi-step reasoning.
Z.ai open foundation model for systems design and long-horizon agent workflows.
Z.ai reasoning model for long-horizon agent workflows and software engineering.
Z.ai reasoning model for complex software engineering and long-horizon agent tasks.
Z.ai native multimodal Flash model for efficient coding and long-horizon tasks.