AI Agent Directory

Alibaba's flagship open-source LLM. 235B MoE (22B active). Multilingual, strong on coding and math. Qwen3-Coder variant matches Claude Code on HumanEval.

DeepSeek's open reasoning model that rivalled o3 at <10% the cost. MIT licensed. 671B MoE. Distilled versions (7B-70B) run on consumer hardware.

Google DeepMind's latest video generation model. Veo 3.1 creates 4K video with native audio — ambient sounds, dialogue, music — all from a single prompt.

Lowest-latency real-time voice AI. Sub-50ms response, 70+ voices, 40+ languages. Used by AI phone agents and live translation services.

Enterprise-grade speech AI. STT with 90%+ accuracy in 30+ languages, TTS, and real-time voice agents. Powers 200K+ developers and Fortune 500 companies.

Mistral's dedicated coding model. 80B parameters, trained on 80+ programming languages. Fill-in-the-middle, instruction following, and test generation. Free ...

The fastest LLM inference API on earth. LPU (Language Processing Unit) hardware runs Llama 4, DeepSeek R1, Gemma 3, and Mixtral at 1000+ tokens/second.

Microsoft's enterprise-grade OpenAI API. GPT-5, o3, DALL-E 3, Whisper with enterprise SLAs, private VNETs, content filtering, and SOC2/HIPAA compliance.