Best Open Source AI Models

Closed-weights API-only models from major labs.

The most capable models from each major lab.

Long chain-of-thought models built for hard math, code, and planning.

Production-grade workhorses with the best speed and cost.

Models specialised for software engineering tasks.

Models that natively understand text, images, and beyond.

Generate still images from a text prompt.

Generate short video clips from a text prompt.

Generate natural-sounding speech audio from text.

All models from OpenAI — GPT, o-series and beyond.

Claude family of models from Anthropic.

Gemini family of models from Google DeepMind.

Open-weights Llama family from Meta AI.

Models from Mistral AI — EU-hosted, multilingual.

Frontier-class open-weights models from DeepSeek.

Grok models from xAI.

Qwen family of open-weights models from Alibaba.

Midjourney text-to-image models.

FLUX image-generation models from Black Forest Labs.

Stable Diffusion image models from Stability AI.

Gen-series text-to-video models from Runway.

Kling text-to-video models from Kuaishou.

Text-to-speech voice models from ElevenLabs.

Sonic low-latency text-to-speech models from Cartesia.