Find the right model.
Or stop choosing.
Compare 51 routed model IDs across 17 providers. Start with text-generator/auto when you want the router to decide.
Auto routenorth_east
Autotext-generator/auto
Semantic routing across providers for a strong quality, speed, and cost balance.
autochatvisiontools
Auto routenorth_east
Auto Fasttext-generator/auto-fast
Low-latency routing for interactive assistants, extraction, and quick agent loops.
autofastchat
Auto routenorth_east
Auto Cheaptext-generator/auto-cheap
Lowest-cost suitable routing for formatting, classification, extraction, and simple edits.
autolow-costchat
Auto routenorth_east
Auto Codetext-generator/auto-code
Routes coding prompts by complexity, from small diffs through architecture and debugging.
autocodetoolsreasoning
Auto routenorth_east
Auto Reasoningtext-generator/auto-reasoning
Automatic model and thinking-level selection for planning, math, and hard decisions.
autoreasoningtools
Auto routenorth_east
Auto Visiontext-generator/auto-vision
Image understanding and OCR routed for latency, resolution, and task complexity.
autovisionocrappnz/auto
App.nz semantic routing across hosted providers with one metered API key.
chatreasoningcodevisiongpt-5.6-luna
Low-cost GPT-5.6 route for high-volume continuation, chat, and tool use.
fastreasoningcodetoolsgpt-5.4
Frontier general reasoning and agentic work.
reasoningcodetoolsvisiongpt-5.4-mini
Efficient reasoning, coding, and tool use.
reasoningcodetoolsvisiongpt-4o
General-purpose multimodal chat model.
chatvisiontoolsgpt-4o-mini
Fast, affordable multimodal model.
fastvisiontoolsclaude-opus-latest
Anthropic flagship for long-horizon reasoning, agentic coding, and knowledge work.
reasoningcodevisiontoolsclaude-opus-5
Latest Claude Opus. Deep reasoning, agentic coding, and 1M-token context.
reasoningcodevisiontoolsclaude-opus-4-8
Previous-generation Claude Opus for reasoning and long-horizon agentic work.
reasoningcodevisiontoolsclaude-sonnet-latest
High-quality coding, analysis, and agentic work at balanced cost.
reasoningcodevisiontoolsclaude-haiku
Fast Claude model for lightweight conversational tasks.
fastchatvisiongemini-2.5-pro
Long-context multimodal reasoning from Google.
reasoningcodevisiontoolsgemini-2.5-flash
Low-latency multimodal model for high-throughput applications.
fastvisiontoolsgrok-4.5
xAI flagship for coding, agentic tasks, and knowledge work with vision and configurable reasoning effort.
reasoningtoolsvisiongrok-4.3
xAI 1M-context reasoning model with tool and vision support at low cost.
reasoningtoolsvisiongrok-4.20-0309-non-reasoning
Fast 1M-context Grok for direct responses and tool-calling workloads.
fasttoolsvisiongrok-build-0.1
xAI coding and build model for software engineering agents.
codeagentictoolsgrok-3-mini
Compact reasoning model for lower-cost tasks.
fastreasoningdeepseek-chat
Cost-efficient general chat and coding model.
fastchatcodedeepseek-reasoner
Reasoning-focused DeepSeek model.
reasoningcodemistral-large-latest
Mistral's flagship general and reasoning model.
chatreasoningtoolsmistral-small-latest
Fast and efficient general-purpose model.
fastchattoolscodestral-latest
Code generation, completion, and repository understanding.
codetoolspixtral-large-latest
Mistral multimodal model for image and document analysis.
visionchatllama-3.3-70b-versatile
Fast Llama inference on Groq.
fastchatcodellama-3.1-8b-instant
Very low-latency chat and extraction.
fastlow-costor/gpt-5.4
Alternate OpenRouter path to GPT-5.4.
reasoningcodefallbackor/claude-sonnet
Alternate OpenRouter path to Claude Sonnet.
reasoningcodefallbackor/claude-opus
Alternate OpenRouter path to Claude Opus.
reasoningfallbackor/gemini-pro
Alternate OpenRouter path to Gemini Pro.
visionreasoningfallbackor/deepseek-chat
Alternate OpenRouter path to DeepSeek.
chatfallbackor/grok-4
Alternate OpenRouter path to Grok.
reasoningfallbackqwen3.5-397b
Large open-weight Qwen model on Together AI.
reasoningcodeopen-weightskimi-k3
Moonshot flagship: 2.8T-parameter open-weights model with 1M context, vision, tool use, and deep reasoning.
reasoningcodevisiontoolsor/kimi-k3
Alternate OpenRouter path to Kimi K3.
reasoningcodefallbackkimi-k2.5
Long-context agentic and coding model.
reasoningcodeopen-weightsminimax-m2.5
MiniMax reasoning and coding model.
reasoningcodeglm-5
Z.AI flagship language and coding model.
reasoningcodetoolsnemotron-ultra
NVIDIA flagship 550B MoE reasoning model served through NIM.
reasoningopen-weightsnemotron-3-super
Mid-size NVIDIA Nemotron MoE for reasoning and agents on NIM.
reasoningtoolsopen-weightsnemotron-3-nano
Low-cost NVIDIA Nemotron MoE for fast chat and extraction on NIM.
fastlow-costopen-weightsnemotron-super-49b
Llama-based Nemotron tuned for reasoning, served through NIM.
reasoningopen-weightsnvidia-gpt-oss-120b
OpenAI open-weights 120B served through NIM.
chatcodeopen-weightsfireworks-llama-70b
Production Llama inference on Fireworks AI.
chatopen-weightshermes-4
Steerable Nous assistant model for tools and agents.
agentictoolsopen-weights