Home Models System Install Pricing Docs Changelog Launch Arcana
6 models available
OpenAI Most Popular

GPT-5.6 Sol

Flagship OpenAI model for complex reasoning, coding, and agentic workflows.

Context
1M
Input
$5/M
Reliability
High
Anthropic Reasoning

Claude Opus 5

Anthropic flagship for demanding reasoning, coding, and visual analysis.

Context
1M
Input
$5/M
Reliability
High
Google Coder

Gemini 3.7 Flash

Fast multimodal Google model for coding, agent workflows, and stepwise reasoning.

Context
1M
Input
$0.38/M
Reliability
High
DeepSeek Reasoning

DeepSeek V4 Pro

Large-scale MoE general-availability release for coding and reasoning.

Context
1M
Input
$0.44/M
Reliability
High
Meta Multimodal

Muse Spark 1.2

Open-weight Meta multimodal agent handling text, images, video, audio, and PDF.

Context
1M
Input
$1.25/M
Reliability
High
Alibaba Reasoning

Qwen3.8 Max

Alibaba flagship multimodal reasoning model in the Qwen3.8 series.

Context
1M
Input
$2/M
Reliability
High
Moonshot AI Reasoning

Kimi K3

2.8T-parameter open multimodal reasoning model for coding and agents.

Context
1M
Input
$3/M
Reliability
High
xAI Reasoning

Grok 4.6

xAI frontier model targeting coding, knowledge work, and STEM tasks.

Context
500K
Input
$2/M
Reliability
High
NVIDIA General

Nemotron 3.5 Lightning

Open MoE from NVIDIA for high-throughput agentic and specialized workloads.

Context
1M
Input
$0.10/M
Reliability
High
Thinking Machines Multimodal

Inkling

Open multimodal MoE from Thinking Machines Lab for reasoning and agents.

Context
1M
Input
$0.95/M
Reliability
High
Poolside Coder

Laguna S 2.1

Poolside coding-agent model for autonomous issue and workflow resolution.

Context
1M
Input
$0.09/M
Reliability
Variable
ByteDance Coder

Seed 2.1 Turbo

Multimodal ByteDance Seed model built for coding and extended agent workflows.

Context
256K
Input
$0.50/M
Reliability
High
Meituan Coder

LongCat 2.0

Sparse Meituan MoE for coding and long-horizon problem solving.

Context
1M
Input
$0.30/M
Reliability
Variable
Upstage General

Solar Pro4

Cost-efficient LLM with a 512K context window for document-heavy agents.

Context
512K
Input
$0.03/M
Reliability
Variable
OpenRouter General

Auto Beta

Task-aware router that classifies requests and sends them to a matching model.

Context
2M
Input
Custom
Reliability
Variable
Alibaba Coder

Qwen3.8 27B

Open dense vision-language model for coding, research, and long agentic runs.

Context
256K
Input
$0.45/M
Reliability
High
Meta Open

Muse Glimmer 30B

Dense open-weight multimodal model distilled for consumer-hardware agents.

Context
128K
Input
$0.35/M
Reliability
High
OpenAI General

GPT-5.6 Luna

Fast, cost-efficient OpenAI model for chat, classification, and lightweight agents.

Context
1M
Input
$0.10/M
Reliability
High
Anthropic Reasoning

Claude Sonnet 5

Anthropic's most capable Sonnet-class model across coding, agents, and professional work.

Context
1M
Input
$2/M
Reliability
High
Anthropic Reasoning

Claude Fable 5

Anthropic frontier model for long-form writing, creative tasks, and narrative reasoning.

Context
1M
Input
$10/M
Reliability
High
xAI Reasoning

Grok 4.5

xAI frontier model with strong coding, knowledge work, and STEM reasoning.

Context
500K
Input
$2/M
Reliability
High
Moonshot AI Reasoning

Kimi K2.6

Long-context open reasoning model for coding and extended agentic workflows.

Context
256K
Input
$0.54/M
Reliability
High
Moonshot AI Coder

Kimi K2.7 Code

Code-specialized Moonshot model with strong reasoning for software engineering tasks.

Context
256K
Input
$0.71/M
Reliability
High
OpenAI General

GPT-5.5

Balanced OpenAI model for everyday coding, reasoning, and agentic work.

Context
1M
Input
$5/M
Reliability
High
OpenAI General

GPT-5

Versatile OpenAI model for general tasks, agents, and multimodal workflows.

Context
400K
Input
$1.25/M
Reliability
High
OpenAI General

GPT-4.1

Reliable general-purpose model with strong tool use and long-context handling.

Context
1M
Input
$2/M
Reliability
High
Google Multimodal

Gemini 2.5 Pro

Google's strongest multimodal model for reasoning, coding, and complex tasks.

Context
1M
Input
$1.25/M
Reliability
High
Google Multimodal

Gemini 2.5 Flash

Fast, efficient Google multimodal model for high-volume agents and apps.

Context
1M
Input
$0.30/M
Reliability
High
DeepSeek Coder

DeepSeek V3.2

General-purpose MoE with strong coding performance and fast inference.

Context
164K
Input
$0.27/M
Reliability
High
DeepSeek Reasoning

DeepSeek R1

High-performance open reasoning model with step-by-step chain-of-thought.

Context
64K
Input
$0.70/M
Reliability
High
Alibaba Coder

Qwen3.7 Flash

Vision-language reasoning model for agents, visual coding, and UI interaction.

Context
1M
Input
$0.03/M
Reliability
High
Alibaba Reasoning

Qwen3.7 Plus

Strong reasoning model in the Qwen3.7 series for complex agentic workflows.

Context
1M
Input
$0.32/M
Reliability
High
Alibaba Coder

Qwen3 Coder

Code-specialized Qwen model for software generation and technical tasks.

Context
1M
Input
$0.65/M
Reliability
High
Mistral General

Mistral Large

Mistral's flagship general-purpose model with strong multilingual reasoning.

Context
131K
Input
$2/M
Reliability
High
Mistral Coder

Codestral 25.08

Mistral's code-specialized model for software engineering and fill-in-the-middle tasks.

Context
256K
Input
$0.30/M
Reliability
High
Mistral General

Mistral Small

Efficient Mistral model for fast, low-cost inference and lightweight agents.

Context
128K
Input
$0.05/M
Reliability
High
Perplexity Reasoning

Sonar Pro

Perplexity search-augmented model with live citations and deep research.

Context
200K
Input
$3/M
Reliability
High
Amazon Multimodal

Nova Pro

Amazon's strongest multimodal model for enterprise agents and document understanding.

Context
300K
Input
$0.80/M
Reliability
High
Cohere General

Command R+

Cohere enterprise model with strong RAG and tool-use capabilities.

Context
128K
Input
$2.50/M
Reliability
High
IBM Coder

Granite 4.1 8B

Compact IBM Granite model for code repair, agents, and trustworthy enterprise use.

Context
131K
Input
$0.05/M
Reliability
High
Nous Research Reasoning

Hermes 4 405B

Open-weight reasoning model from Nous Research for uncensored agentic work.

Context
131K
Input
$1/M
Reliability
Variable
Sakana AI Reasoning

Fugu Ultra

1B-parameter continuous-thinking model from Sakana AI for deep reasoning.

Context
1M
Input
$5/M
Reliability
Variable
Liquid AI Open

LFM 2.5 2.6B

Compact free reasoning model from Liquid AI for agents and long-context tasks.

Context
128K
Input
Free
Reliability
Variable
Aion Labs General

Aion 3.0

Multi-model roleplaying and storytelling system built on the GLM family.

Context
131K
Input
$3/M
Reliability
Variable
MiniMax Multimodal

MiniMax M3

Multimodal agent model with strong long-context and agentic capabilities.

Context
1M
Input
$0.30/M
Reliability
High
StepFun Reasoning

Step 3.7 Flash

StepFun fast reasoning model for coding, math, and long-horizon agents.

Context
256K
Input
$0.20/M
Reliability
High
Z AI Reasoning

GLM 5.2

Z AI flagship multimodal reasoning model with 1M context support.

Context
1M
Input
$0.31/M
Reliability
High
Tencent Reasoning

Hunyuan HY3

295B-parameter MoE from Tencent built for reasoning, coding, and math.

Context
256K
Input
$0.13/M
Reliability
Variable
ByteDance Coder

Seed 2.0 Lite

Lightweight ByteDance Seed model for agentic coding and multilingual environments.

Context
256K
Input
$0.25/M
Reliability
High
NVIDIA Open

Nemotron 3 Super

120B-parameter open MoE from NVIDIA for agentic and specialized workloads.

Context
1M
Input
$0.09/M
Reliability
High

No models match .