DeepSeek V4.1 Flash 1017 is an open-weight multimodal model optimized for efficient reasoning, coding, and agentic workflows with a 1-million-token context window.

GLM 5.3 Flash is a fast, cost-efficient multimodal model built for long-context reasoning and scalable workflows.

Kimi K3.1 Pro is Moonshot's 3T multimodal agentic model for coding with 1M context.

GLM-5.3 is a flagship open-weight model optimized for advanced coding, agentic workflows, and long-context reasoning.

Qwen3.8 27B is a compact open-weight multimodal model optimized for efficient reasoning, coding, and agentic workflows on local hardware.

DeepSeek V4 Flash 0731 is an efficient MoE model optimized for fast reasoning, coding, and enhanced agentic workflows.

Kimi K3 is a frontier open-weight multimodal model built for long-horizon reasoning, coding, and advanced agentic workflows.

GLM-5.2 is a flagship model built for long-horizon agentic tasks with stable 1M-token context and advanced coding capabilities.

Claude Haiku 5.5 is Anthropic’s fastest and most cost-efficient small model, built for high-volume workloads that require responsive performance, low operating costs, and reliable handling of repetitive tasks and lightweight agentic workflows.

Claude Sonnet 5.5 is Anthropic’s mid-tier AI model, designed to deliver near-flagship coding and knowledge-work performance with faster execution, lower token usage, and reduced cost for everyday professional tasks.

Claude Opus 5.5 is Anthropic’s most capable Opus model, built for high-end coding and long-running agent workflows.

Claude Opus 5 is Anthropic’s most advanced Opus model, built for long-running agents, complex coding, and high-level professional workflows.

Claude Sonnet 5 is a high-performance model optimized for coding, agent workflows, and everyday professional tasks with strong efficiency and scalability.

Claude Opus 4.8 is a top-tier model built for sustained autonomous work, excelling in coding, agents, and complex enterprise tasks.

Gemini 4 Argon is Google’s flagship reasoning model, built for complex software engineering, enterprise problem-solving, and long-running autonomous workflows.

DeepSeek V4.1 Flash is an efficient multimodal MoE model optimized for fast agentic reasoning and coding workflows.

Gemini 3.8 Flash is a fast, cost-efficient multimodal model built for advanced coding and agentic workflows.

Mistral OCR 3 is an efficient document AI model for structured text extraction and LLM-ready document conversion.

Gemini 3.7 Flash is a high-efficiency model delivering strong agentic capabilities, improved coding performance, and reliable multi-step execution.

Gemini 3.6 Flash is a high-efficiency model optimized for multi-step workflows, coding, and improved reasoning performance.

Gemini 3.5 Flash Lite is a fast, low-cost multimodal model optimized for high-volume, low-latency workloads.

GPT-5.6 Sol is the flagship model of the GPT-5.6 family, built for advanced reasoning, deep research, and complex multi-step execution.

GPT-5.6 Terra is a balanced mid-tier model optimized for professional work, combining strong reasoning, coding ability, and efficiency.

GPT-5.6 Luna is a fast, cost-efficient model optimized for high-volume tasks, real-time interactions, and lightweight reasoning.

Gemma 4 31B IT is a multimodal instruction-tuned model from Google DeepMind designed for strong reasoning, coding, and vision-language tasks.

Gemini 3.1 Flash Lite is a highly efficient multimodal model optimized for ultra-low latency and cost-sensitive, high-volume workloads.

Gemini 3.8 Flash Lite TTS is a fast, cost-efficient text-to-speech model built for high-throughput production workloads and conversational agents.

Gemini 3.8 Flash TTS is a text-to-speech model built for studio-grade voice fidelity, expressive acting, and natural speech.

MiniMax M3.1 Flash - the speed-optimized model, delivering low-latency, cost-efficient inference while preserving strong reasoning, coding, and long-context handling for high-volume workloads.

Voxtral Small Transcribe 3 is Mistral AI’s multilingual speech-to-text model, designed for batch processing with speaker diarization and context-aware transcription.