70+ models from 20+ providers

Auto

Auto classifies each prompt and routes it to a strong model for the task.

GPT-6 Astra

GPT-6 Astra is a frontier reasoning model built for advanced coding, multistep tool use, and complex professional work.

GPT-6.1 Sol

GPT-6.1 Sol is a reasoning model with near-Astra performance on agentic coding, computer use, and professional work.

GPT-6.1 Terra

GPT-6.1 Terra is a fast, efficient reasoning model with detailed visual reasoning, precise computer use, and clear writing.

GPT-6 Sol

GPT-6 Sol is a reasoning model for complex professional work, including coding, at lower cost than GPT-6 Astra.

GPT-6.1 Sol Pro

GPT-6.1 Sol Pro runs GPT-6.1 Sol in Pro mode for the hardest problems, using more reasoning and taking longer to answer.

GPT-6 Sol Pro

GPT-6 Sol Pro runs GPT-6 Sol in Pro mode for the hardest problems, using more reasoning and taking longer to answer.

GPT-6 Luna

GPT-6 Luna is an efficient reasoning model for high-volume, clearly scoped tasks like summarizing or extracting information.

GPT-5.6 Terra

GPT-5.6 Terra balances reasoning quality and efficiency.

GPT-4o mini

GPT-4o mini is a small, fast multimodal model for everyday chat. Supports web search.

gpt-oss-120b

gpt-oss-120b is a stronger open-weight 120B model for higher quality responses.

gpt-oss-20b

gpt-oss-20b is a fast open-weight 20B model for everyday chat.

Claude Fable 5.1

Claude Fable 5.1 is tuned for long-context creative and analytical work, with more efficient reasoning and improved coding accuracy over Fable 5. Supports reasoning and web search.

Claude Fable 5

Claude Fable 5 is tuned for long-context creative and analytical work. Supports reasoning and web search.

Claude Opus 5.5

Claude Opus 5.5 excels at agentic coding, large-scale code migrations, and complex professional work. Supports reasoning and web search.

Claude Opus 5

Claude Opus 5 excels at coding, debugging, and novel problem-solving. Supports reasoning and web search.

Claude Sonnet 5.5

Claude Sonnet 5.5 is fast and capable for everyday coding, agentic tasks, and professional work. Supports reasoning and web search.

Claude Sonnet 5

Claude Sonnet 5 is strong for everyday coding, writing, and analysis. Supports reasoning and web search.

Claude Haiku 5.5

Claude Haiku 5.5 is the fastest option for high-volume tasks like classification, extraction, and routing. Supports reasoning and web search.

Claude Haiku 4.5

Claude Haiku 4.5 is fast for quick answers and light coding. Supports web search.

Gemini 4 Argon

Gemini 4 Argon is a flagship reasoning model for complex software engineering, scientific problem solving, and long multi-step agent work. Supports reasoning and web search.

Gemini 3.8 Flash

Gemini 3.8 Flash is a fast general-purpose model built for coding, agentic tasks, and multi-step reasoning. Supports reasoning and web search.

Gemini 3.7 Flash

Gemini 3.7 Flash is a workhorse model built for coding and agentic tasks, with strong web app generation and document understanding. Supports reasoning and web search.

Gemini 3.5 Flash Lite

Gemini 3.5 Flash Lite is the fastest 3.5-class model for high-throughput chat and light agent work. Supports reasoning and web search.

Gemini 3.1 Pro

Gemini 3.1 Pro is built for deeper analysis and long context. Supports reasoning and web search.

Gemini 3.1 Flash Lite

Gemini 3.1 Flash Lite is lightweight for snappy chat and light reasoning. Supports web search.

Gemma 4 31B

Gemma 4 31B is a capable dense model for general chat. Supports web search.

Gemma 4 31B 🎁

Gemma 4 31B is a denser open model for stronger general chat. Supports web search.

Gemma 4 26B A4B

Gemma 4 26B A4B is an efficient mixture-of-experts model for general chat. Supports web search.

Gemma 4 26B A4B 🎁

Gemma 4 26B A4B is a lightweight mixture-of-experts model. Supports web search.

Grok 4.7

Grok 4.7 is the flagship for coding, knowledge work, and long tasks. Supports reasoning, image input, and web search.

Grok 4.6

Grok 4.6 is a strong model for current knowledge, reasoning, and web search.

DeepSeek V4.1 Flash 1017

DeepSeek V4.1 Flash 1017 is a fast, cost-efficient model for coding, reasoning, and long agentic workflows, with a 1M-token context. Supports web search.

DeepSeek V4 Flash 0731

DeepSeek V4 Flash is for quick reasoning and coding. Supports web search.

DeepSeek V4 Pro 0813

DeepSeek V4 Pro is the flagship for coding and hard reasoning. Supports web search.

☁️ DeepSeek R1 (⚗️ Qwen 32B)

DeepSeek R1 (Qwen 32B distill) is reasoning-focused for multi-step problem solving.

Kimi K3.1 Pro

Kimi K3.1 Pro is a 3T multimodal reasoning model with 1M context, strong at coding, 3D creation and financial research.

Kimi K3

Kimi K3 is a 2.8T multimodal reasoning model with 1M context.

Kimi K2.6

Kimi K2.6 offers long-context reasoning. Supports web search.

Kimi K2.7 Code

Kimi K2.7 Code is a software engineering specialist with reasoning and web search.

MiniMax-M3.1-Flash 🎁

MiniMax-M3.1-Flash is a fast reasoning model with solid multilingual performance. Supports web search.

MiniMax M3

MiniMax M3 is a reasoning model with solid multilingual performance. Supports web search.

GLM-5.3

GLM-5.3 is a frontier coding and agentic model with strong long-horizon task handling. Supports reasoning and web search.

GLM-5.3 Flash

GLM-5.3 Flash is a fast, lightweight coding and agentic model. Supports reasoning and web search.

GLM-5.2

GLM-5.2 is a bilingual flagship with reasoning and web search.

Deep Max

Deep Max runs maximum-effort agentic web search and synthesis for hard research questions.

Answer (Pro)

Answer Pro returns higher-quality, source-rich web answers.

Answer

Answer returns web-grounded responses optimized for retrieval quality.

Sonar Reasoning Pro

Sonar Reasoning Pro does multi-step research with web search and chain-of-thought.

Sonar Pro

Sonar Pro goes deeper on research-style answers with built-in web search.

Sonar

Sonar gives grounded answers with built-in web search.

Sonar Deep Research

Sonar Deep Research runs long-form, multi-source investigations.

Mistral Large 4

Mistral Large 4 is a multimodal flagship with a 1M-token context for complex reasoning, vision, and tool use.

Mistral Large 3

Mistral Large 3 is the flagship for complex reasoning and multilingual work.

Mistral Small 4

Mistral Small 4 is efficient for everyday chat, coding, and quick reasoning.

Codestral

Codestral is a code specialist for generation, completion, and debugging.

Command A

Command A is built for enterprise-style chat, RAG, and tool use.

Inkling

Inkling is a multimodal MoE model (41B active, 975B total) for reasoning, coding, and multilingual chat, with 1M context. Supports web search.

Inkling 2 Small

Inkling 2 Small is a multimodal MoE model (17B active, 594B total) for agentic coding, vision, knowledge work, and reasoning, with adjustable thinking effort and 256K context.

Mercury 3

Mercury 3 is a fast reasoning diffusion model with a 1M context window.

Muse Wonder

Muse Wonder is a frontier multimodal reasoning model for personal and professional agents, with Active Memory, Active Focus, and long-running coding.

Muse Spark 1.3

Muse Spark 1.3 is a multimodal reasoning model for coding and agentic work, with 1M context and more efficient tool use than Muse Spark 1.2.

Muse Spark 1.2

Muse Spark 1.2 is a multimodal reasoning model for coding and agentic work, with 1M context.

Muse Spark Open

Muse Spark Open is an open-weights multimodal reasoning model for coding and agentic work, with 1M context.

Muse Spark 1.1

Muse Spark 1.1 is an earlier multimodal reasoning checkpoint for coding and long-context work.

Llama 4 Maverick

Llama 4 Maverick is a higher-capacity multimodal model. Supports web search.

Llama 4 Scout

Llama 4 Scout is a fast multimodal model. Supports web search.

Llama 3.3 70B

Llama 3.3 70B has strong instruction following. Supports web search.

☁️ Llama 3.1 70B

Llama 3.1 70B is a reliable open model for general chat.

MiMo-V2.5-Pro

MiMo-V2.5-Pro is strong at reasoning and web search for general work.

MiMo-V2.5

MiMo-V2.5 improves reasoning and supports web search.

Qwen3.8 Flash

Qwen3.8 Flash is a fast, cost-efficient model for reasoning and coding. Supports web search.

Qwen3.7 Max

Qwen3.7 Max is the flagship for maximum multilingual quality with reasoning and web search.

Qwen3.7 Plus

Qwen3.7 Plus is strong at reasoning, coding, and multilingual work. Supports web search.

Nemotron 3.5 Super VL

Nemotron 3.5 Super VL is a multimodal reasoning model that understands text, images, documents, and video, built for agentic workflows with a 1M context window.

Nemotron 3.5 Super VL 🎁

Nemotron 3.5 Lightning

Nemotron 3.5 Lightning is NVIDIA's fast, low-cost mixture-of-experts model for high-throughput agentic workloads, with a 262K context window.

Nemotron 3.5 Lightning 🎁

Nemotron 3 Ultra

Nemotron 3 Ultra is a massive mixture-of-experts model for hard tasks with reasoning and web search.

Nemotron 3 Ultra 🎁

Nemotron 3 Ultra is a large mixture-of-experts model with reasoning and web search.

Phi-4

Phi-4 is a compact model for chat and light coding. Supports web search.

Nova Pro

Nova Pro is for higher-quality general work. Supports web search.

Nova Lite

Nova Lite is for fast everyday chat. Supports web search.

Nova Micro

Nova Micro is an ultra-small model for simple prompts. Supports web search.