DeepSeek-R1https://deepseek.comOpen reasoning model. Claude 3.5 Sonnethttps://www.anthropic.com/claudePrevious-gen Claude, still widely deployed. GPT-4ohttps://platform.openai.com/docs/models/gpt-4oOpenAI multimodal flagship. Llama 3.1 405Bhttps://llama.meta.comMeta's largest open model. DeepSeek-V3https://deepseek.comMoE open-weights flagship. Qwen2.5 72Bhttps://qwen.aiAlibaba's strong open model. Llama 3.3 70Bhttps://llama.meta.comMeta's flagship open-weights model. Mistral Large 2https://mistral.aiMistral's dense flagship. GPT-4o minihttps://platform.openai.com/docs/models/gpt-4o-miniCheap GPT-4o-class small model. Mixtral 8×22Bhttps://mistral.aiMistral's open MoE. Claude Sonnet 4.5https://www.anthropic.com/claude/sonnetAnthropic's coding-focused frontier model. Command R+https://cohere.com/commandCohere RAG-tuned model. Kronoshttps://github.com/shiyu-coder/KronosFoundation model for financial K-line data. MiniMax M2https://github.com/MiniMax-AI/MiniMax-M2MiniMax's agent model. Kimi K2.5https://github.com/MoonshotAI/Kimi-K2.5Moonshot's agentic K2.5 stack. Claude Fable 5https://www.anthropic.com/claudeAnthropic Fable generation. Claude Fable 5.1https://www.anthropic.com/claudeAnthropic Fable frontier. Claude Opus 4.8https://www.anthropic.com/claudeOpus 4.x flagship. Claude Opus 5https://www.anthropic.com/claudeAnthropic Opus tier. Claude Sonnet 5https://www.anthropic.com/claudeSonnet 5 mid tier. DeepSeek-V4 Flashhttps://deepseek.comDeepSeek V4 fast variant. DeepSeek-V4 Prohttps://deepseek.comDeepSeek V4 reasoning flagship. Eleven v3https://elevenlabs.ioExpressive TTS. FLUX 1.1 Prohttps://bfl.aiBlack Forest Labs image generation. FLUX.1 [dev]https://github.com/black-forest-labs/fluxOpen-weights FLUX image model. Gemini 2.5 Flashhttps://deepmind.google/models/gemini/flash/Fast multimodal Gemini; free AI Studio tier. Gemini 2.5 Prohttps://deepmind.google/models/gemini/pro/Google's top reasoning model; video + audio input. Gemini 3.1 Prohttps://deepmind.google/models/gemini/Gemini 3.1 reasoning pro. Gemini 3.6 Flashhttps://deepmind.google/models/gemini/Gemini 3.6 fast multimodal. Gemini 3.7 Flashhttps://deepmind.google/models/gemini/Gemini 3.7 fast multimodal. GLM-4.6https://z.aiZ.ai's open coding model. GLM-5.3https://z.aiZ.ai GLM 5.3 flagship. GLM-5.3 Flashhttps://z.aiGLM 5.3 fast variant. GPT-5.3 Codexhttps://platform.openai.comGPT-5.3 coding-specialized. GPT-5.4https://platform.openai.comGPT-5.4 generation. GPT-5.5https://platform.openai.comGPT-5.5 flagship. GPT-5.6https://platform.openai.comOpenAI frontier flagship family. GPT-5.6 Solhttps://platform.openai.comGPT-5.6 reasoning variant. Grok 4https://x.aixAI Grok 4. Grok 4.6https://x.aixAI Grok frontier. HunyuanVideohttps://github.com/Tencent/HunyuanVideoTencent open video generation. K2 Thinkinghttps://kimi.moonshot.cnKimi K2 reasoning variant. Kimi K2https://kimi.moonshot.cnMoonshot's open agentic model. Kimi K3https://kimi.moonshot.cnMoonshot's K3 agentic flagship. Kling 2https://klingai.comKuaishou video generation. MiniMax M2https://www.minimax.ioMiniMax's open agent model. Morph V3https://morphllm.comMorph's fast apply/edit model. Nemotron Ultrahttps://build.nvidia.comNVIDIA's open reasoning model. Qwen 3.7 Maxhttps://qwen.aiQwen 3.7 generation. Qwen 3.8 Maxhttps://qwen.aiAlibaba's Qwen 3.8 flagship. Sora 2https://openai.com/soraOpenAI video generation. Veo 3https://deepmind.google/models/veo/Google video generation. Wan 2.5https://github.com/Wan-VideoAlibaba open video model. Whisper large-v3https://github.com/openai/whisperOpenAI open speech-to-text.