GPT-6.1 Sol is a model from OpenAI for complex coding, computer use, and professional work, offering near-Astra performance at a lower cost.

Claude Sonnet 5.5 is a fast Claude model from Anthropic for everyday coding, agents, and knowledge work.

Claude Opus 5.5 is a Claude model for long-running agentic coding and knowledge work.

GPT-6 Luna is OpenAI's most efficient model for focused, high-volume tasks.

GPT-6 Sol is an OpenAI model for complex coding and agentic workflows.

MiMo-V2.6-Flash is a multimodal model from Xiaomi for coding agents and long-context automation.

MiMo-V2.6-Pro is Xiaomi's Pro model in the MiMo line, built for multimodal coding agents and long-context automation.

Grok 4.7 is xAI's frontier model for long-running agents, coding, knowledge work, and visual projects.

GLM-5.3-FlashX is a high-speed serving option from Z.ai for coding and agent workflows.

Qwen3.8 Omni Flash is an omni model from Alibaba's Qwen line for text, vision, audio, and multimodal agent tasks.

Step 5 Preview is StepFun's next-generation flagship base model for coding and professional knowledge work.

DeepSeek V4 Flash Vision Exp is a model in the deepseek-flash line, described by the maker as a DeepSeek V4.1 Flash model for reasoning and agentic coding.

DeepSeek V4.1 Flash is a model for reasoning and agentic coding, part of the deepseek-flash line.

Mercury 2.5 is a reasoning LLM and diffusion LLM (dLLM) from Inception.

GPT-6 Astra is OpenAI's most capable model for complex reasoning, coding, computer use, research, and document creation.

Gemini 3.8 Flash is Google's most intelligent Flash model, engineered for long-horizon software engineering, autonomous agents, and complex enterprise workflows.

Muse Spark 1.3 is a multimodal reasoning model from Meta for long-running agentic, multi-agent, and coding workflows.

Claude Fable 5.1 is a Claude model for demanding reasoning and long-horizon agentic work.

GLM-5.3-Flash is a native multimodal GLM model from Z.ai (Zhipu, GLM) designed for efficient coding and long-horizon agent tasks.

Qwen3.8 Flash is a vision-language model from Alibaba's Qwen line, designed for visual reasoning, documents, and agent tasks.

GLM-5.3 is the flagship model in the GLM line from Z.ai (Zhipu).

Gemini 3.7 Flash is a high-efficiency model from Google for agentic workflows, coding, and multimodal reasoning.

DeepSeek V4 Pro is a snapshot in the deepseek-thinking line, released on 2026-08-12.

Grok 4.6 is xAI's frontier model for long-running agents, coding, knowledge work, and visual projects.

Nemotron 3.5 Lightning 30B A3B is a fast NVIDIA Nemotron MoE for reliable agentic tasks across enterprise workloads.

Solar Pro 4 is Upstage's flagship model, specialized for agentic use.

Muse Spark 1.2 is a coding-focused update to Muse Spark 1.1 from Meta.

Qwen3.8 Max is a 2.4-trillion-parameter mixture-of-experts flagship from Alibaba's Qwen line, aimed at coding, professional work, multimodal understanding, and long-horizon agentic workflows.

Sakana Namazu is a Japanese-specialized reasoning model based on Kimi K2.6, tuned for Japanese language, culture, and business workflows.

Claude Opus 5 is Anthropic's strongest Claude Opus model for coding, agents, and professional work.

Gemini 3.5 Flash Lite is a fast model from Google that balances multimodal reasoning, tool use, and cost.

Gemini 3.6 Flash is a fast Gemini model that balances multimodal reasoning, tool use, and cost.

Laguna S 2.1 is an agentic coding model from Poolside in the XS size class for local deployment.

Kimi K3 is a multimodal model from Moonshot AI (Kimi) with a 1M token context and toggleable max-effort thinking, designed for long-horizon agent work.

Qwen3.7 Flash is a lightweight multimodal Qwen model for high-throughput text, image, and video tasks.

GPT-5.6 is a frontier model from OpenAI for complex professional work, coding, and agentic workflows.

GPT-5.6 Luna is a cost-efficient model from OpenAI for fast, high-volume workloads.

GPT-5.6 Sol is a frontier model from OpenAI for complex professional work, coding, and agentic workflows.

GPT-5.6 Terra is a balanced model from OpenAI for capable, cost-efficient everyday work.

Grok 4.5 is xAI's model for chat, coding, agentic tools, and lower hallucination risk.

Laguna XS 2.1 is an agentic coding model from Poolside in the XS size class, designed for local deployment.

Meituan LongCat-2.0 is a reasoning model with tool calling and a 1M-token context window.

Claude Sonnet 5 is an everyday Claude agent model for coding, planning, browsing, and general work.

Seed 2.1 Pro is ByteDance's flagship model in the Seed line, designed for complex multimodal reasoning, coding, and agents.

Seed 2.1 Turbo is a faster variant of ByteDance's Seed 2.1 model, designed for multimodal reasoning and latency-sensitive agent workflows.

Fugu is a multi-agent model for routing expert agents across complex analytical tasks.

Fugu Ultra is a multi-agent model from Sakana AI, designed for hard research, analysis, and competitions.

GLM-5.2 is an open flagship model from Z.ai (Zhipu) for long-horizon coding agents and million-token context work.

Kimi K2.7 Code HighSpeed is a coding-focused model from Moonshot AI, designed for long-horizon repository work with less overthinking.

Claude Fable 5 is a Claude model for creative writing, analysis, and controlled agent workflows.

Nemotron 3 Ultra 550B A55B is the largest model in the Nemotron 3 line, designed for maximum open-weight reasoning and agent accuracy.

Qwen3.7 Plus is a multimodal model from Alibaba's Qwen line, positioned as a workhorse for long-context agents, visual inputs, and coding.

MiniMax-M3 is a multimodal model for long-context coding, perception, and agent planning.

Step 3.7 Flash is a newer StepFun flash model designed for faster agents, coding, and multimodal prompts.

Claude Opus 4.8 is Anthropic's top Claude Opus tier, designed for the hardest reasoning, coding, and long-horizon agents.

Qwen3.7 Max is the first version of the Qwen line in this registry.

Command A Plus is Cohere's stronger command model for multilingual agents and enterprise workflows.

Gemini 3.5 Flash is a fast model in the gemini-flash line, balancing multimodal reasoning, tool use, and cost.

Mistral Medium is a balanced model from Mistral AI for enterprise assistants, multilingual work, and tools.

Mistral Medium 3.5 is a balanced Mistral model for enterprise assistants, multilingual work, and tools.

Laguna M.1 is Poolside's open-weight model for agentic coding and long-hizon work.

Nemotron 3 Nano Omni is an open omni model from NVIDIA that combines reasoning with text, vision, and audio.

Qwen3.6 Flash is a vision-language model from Alibaba's Qwen line for visual reasoning, documents, and agent tasks.

GPT-5.5 is OpenAI's default frontier model for coding, computer use, research, and knowledge work.

GPT-5.5 Pro is the highest-accuracy tier in the GPT-5.5 line, designed for slower, precision-heavy reasoning and coding.

MiMo-V2.5 is an open model from Xiaomi for multimodal coding agents and long-context automation.

MiMo-V2.5-Pro is Xiaomi's Pro tier in the MiMo line, built for multimodal reasoning and coding-agent execution.

Kimi K2.6 is a multimodal model from Moonshot AI, positioned as a workhorse for agent loops, coding tasks, and visual context.

Grok 4.3 is xAI's model for chat, coding, agentic tools, and lower hallucination risk.

Grok Build 0.1 is a coding model from xAI, tuned for agentic engineering and iterative edits.

Claude Opus 4.7 is Anthropic's stronger Opus tier for advanced software work and high-stakes reasoning.

Muse Spark 1.1 is a natively multimodal reasoning model from Meta.

GLM-5.1 is a coding model from Z.ai (Zhipu) for agentic engineering, terminals, and repository generation.

Gemma 4 26B A4B IT is an open instruction model from Google for efficient chat and self-hosted deployments.

Gemma 4 31B IT is Google's largest instruction-tuned model in the Gemma line, designed for open, self-hosted chat and reasoning.

Step 3.5 Flash 2603 is a flash model from StepFun for efficient multimodal reasoning, coding, and tool use, according to the maker.

GLM-5V-Turbo is a vision model from Z.ai (Zhipu, GLM) for screenshots, documents, and multimodal agent tasks.

Trinity Large Thinking is a reasoning-optimized 398B MoE agent model from Arcee AI, designed for extended thinking in long-horizon and multi-turn tool use.

Mercury Edit 2 is a code editing dLLM from Inception, designed for autocomplete (FIM) and next-edit suggestions.

MiMo-V2-Omni is a legacy model from Xiaomi's mimo line, retained for compatibility with older integrations.

MiMo-V2-Pro is a model from Xiaomi for multimodal agents, reasoning, and code tasks.

MiniMax-M2.7 is the open flagship model from MiniMax for coding agents, office automation, and complex environments.

GPT-5.4 mini is a small model from OpenAI for coding subagents, quick tool use, and high-volume work.

GPT-5.4 nano is OpenAI's cheapest GPT-5.4 lane, designed for simple routing, extraction, and bulk automation.

GLM-5-Turbo is a faster lane in the GLM-5 line, designed for coding agents that need lower latency.

Mistral Small is an efficient model from Mistral AI for fast chat, extraction, and production assistants.

Mistral Small 4 is a production model from Mistral AI for chat, extraction, and cost-sensitive agents.

Nemotron 3 Super is the middle tier in NVIDIA's Nemotron line, designed for collaborative agents and high-volume reasoning workloads.

Grok 4.20 (Non-Reasoning) is a model from xAI for agentic tool use, reasoning, coding, and live assistance.

Grok 4.20 Multi-Agent is a Grok model from xAI, described for agentic tool use, reasoning, coding, and live assistance.

GPT-5.4 is an agent-ready GPT for coding and computer-use workflows at a lower cost, according to OpenAI.

GPT-5.4 Pro is a model from OpenAI for demanding professional reasoning and agent tasks.

Gemini 3.1 Flash Lite Preview is a legacy model in the gemini-flash-lite line, retained by Google for compatibility with older integrations.

GPT-5.3 Chat is a chat-tuned GPT model from OpenAI for conversational assistance, writing, and tool workflows.

Mercury 2 is a reasoning model from Inception for deliberate analysis, multi-step problem solving, and tool use.

Gemini 3.1 Pro Preview is a reasoning-first Gemini preview from Google for agentic coding and complex problem solving.

Sarvam-105B is a flagship Indian-language reasoning model for enterprise multilingual applications.

Sarvam-30B is an efficient Indian-language reasoning model for chat, coding, and multilingual work, as described by Sarvam AI.

Claude Sonnet 4.6 is Anthropic's workhorse model for coding agents, careful analysis, and production cost control.

Tiny Aya Earth is a compact multilingual model from Cohere, specialized for West Asian and African languages.

Tiny Aya Fire is a compact multilingual model from Cohere, specialized for South Asian languages.

Compact multilingual model balanced across 70 languages and regions.

Tiny Aya Water is a compact multilingual model from Cohere, specialized for European and Asia-Pacific languages.

ByteDance Seed 2.0 Code is a coding model for multimodal software engineering and long-running agents, as described by the maker.

Seed 2.0 Lite is a cost-efficient model from ByteDance's Seed line, designed for production chat, analysis, and structured generation.

Seed 2.0 Mini is a lightweight ByteDance Seed 2.0 model for low-latency multimodal reasoning and high-volume tasks.

ByteDance Seed 2.0 Pro is the flagship model in the Seed line, designed for complex multimodal reasoning and long-horizon agent workflows.

GLM-5 is a general GLM flagship for coding, analysis, and tool-heavy engineering workflows.

MiniMax-M2.5 is a coding model from MiniMax for agent workflows, office edits, and automation, according to the maker.

GPT-5.3 Codex is a coding-optimized GPT model for repository edits, reviews, and agentic software work.

GPT-5.3 Codex Spark is a coding-optimized GPT model from OpenAI for repository edits, reviews, and agentic software work.

Claude Opus 4.6 is Anthropic's high-end model for difficult coding, planning, and slower expert reasoning.

Voxtral Mini is a speech transcription model from Mistral AI, built for accurate audio-to-text and captioning workflows.

Step 3.5 Flash is StepFun's flash lane for quick multimodal reasoning and coding assistance.

GLM-4.7-Flash is a budget model in the GLM line for fast coding help, routing, and everyday automation, as described by Z.ai.

GLM-4.7-FlashX is an efficient GLM model for fast reasoning, coding, and agent workflows.

Jamba Mini is AI21 Labs' efficient, lightweight hybrid SSM-Transformer model for a wide range of tasks.

Solar Pro 3 is the flagship model in the solar-pro line, designed for demanding analysis, coding, and production agent workflows.

MiniMax-M2.1 is an agent model from MiniMax for practical coding and productivity tasks.

Gemini 3 Flash Preview is a new model in Google's gemini-flash line.

GPT-5.2 Chat is a chat-tuned GPT model from OpenAI for conversational assistance, writing, and tool workflows.

GPT-5.2 Pro is a higher-accuracy variant of GPT-5.2 for tougher reasoning and review workflows.

Devstral 2 is Mistral's coding-agent model for repository work, terminal tasks, and software fixes.

Devstral Small 2 is a legacy model from Mistral AI, retained for compatibility with older integrations.

GLM-4.6V-Flash is a lightweight vision model from Z.ai for visual reasoning, documents, and multimodal agents.

Nova 2 Pro is a multimodal reasoning model from Amazon for visual analysis, planning, and tool use.

Nova 2 Lite is a multimodal reasoning model from Amazon for visual analysis, planning, and tool use.

Claude Opus 4.5 is Anthropic's flagship model for deep reasoning, coding, and long-horizon agents.

GPT-5.1 is a model in OpenAI's GPT line, described as a sharper GPT-5 generation for coding, product work, and tool-assisted tasks.

Nemotron Nano 12B v2 VL is a multimodal model from NVIDIA for visual reasoning and agentic AI workflows.

MiniMax-M2 is an open model from MiniMax, built for coding agents and tool-heavy workflows.

Claude Haiku 4.5 is Anthropic's fast lane model for lightweight agents, office tasks, and responsive chat.

Gemini 2.5 Computer Use Preview 10-2025 is a specialized Gemini 2.5 model for browser-control agents that automate UI tasks.

GPT-5 Pro is a higher-accuracy tier in the GPT-5 line, intended for tough analysis, coding reviews, and planning.

Granite-4.0-H-Small is an open-weight hybrid model from IBM for enterprise chat, coding, retrieval-augmented generation, and tool-calling workloads.

Ling-1T is an open-weight instruction model from Ant Group for adaptable chat and self-hosted production workloads.

Ring-1T is a reasoning model from Ant Group for deliberate analysis, multi-step problem solving, and tool use.

Claude Sonnet 4.5 is a balanced Claude model for coding, analysis, agent workflows, and cost control, per Anthropic.

nvidia-nemotron-nano-9b-v2 is a compact Nemotron model from NVIDIA for efficient reasoning and deployable AI agents.

Mistral Medium 3.1 is a Mistral AI model for multilingual chat, reasoning, and tool-assisted workflows.

GPT-5 is OpenAI's original workhorse model for reasoning, coding, writing, and tool workflows.

GPT-5 Mini is a small model from OpenAI for responsive agents, coding help, and everyday automation.

GPT-5 Nano is a tiny model in the GPT-5 lane for routing, extraction, classification, and bulk jobs.

GLM-4.5-Air is a lighter variant of GLM-4.5 from Z.ai (Zhipu, GLM).

GLM-4.5-Flash is an efficient GLM model from Z.ai (Zhipu, GLM) for fast reasoning, coding, and agent workflows.

Voxtral Small is an instruct model from Mistral AI with native audio input for speech understanding and tool use.

Devstral Medium is a legacy model in the devstral line, retained for compatibility with older integrations.

Devstral Small is a legacy model from Mistral AI, retained for compatibility with older integrations.

Jamba Large is AI21 Labs' hybrid SSM-Transformer long-context model, built for enterprise agents and grounded generation.

Mistral Small 3.2 is an efficient Mistral model for fast chat, extraction, and production assistants.

Gemini 2.5 Flash is a fast multimodal model for applications where latency and price matter.

Gemini 2.5 Flash-Lite is a lean version of Gemini 2.5 for cheap multimodal traffic and quick agents.

Gemini 2.5 Pro is Google's reasoning model for coding, math, and multimodal analysis.

o3-pro is a high-effort tier for difficult technical reasoning and careful answers.