Meta AI is the AI division of Meta, known for the open-source Llama model family with over 350 million downloads. Open weights, SAM segmentation and AI Studio span open research to multimodal applications.
TAG COLLECTION
Open Models collection
Explore providers, articles, and companies related to Open Models for practical research and comparison.
Providers
Ollama runs local models with 1 command, exposing an OpenAI-compatible endpoint on port 11434 where migration changes 1 base_url, covering 100 or more open models with Modelfile customisation across macOS, Linux and Windows; the gaps are high-concurrency throughput and production orchestration.
Groq serves ultra-low-latency inference on custom LPU silicon for mainstream open-weight models through an OpenAI-style API where migration usually changes 2 configuration points, with a free tier and per-token paid plans for voice and realtime chat; the gaps are missing closed flagships and capacity limits.
ModelScope hosts tens of thousands of open models and datasets filterable by task and licence, offering free-compute notebooks, repository versioning with 1 command pulls and inference APIs plus 1 step Alibaba Cloud deployment; compute queues, licence variance and average overseas access are the trade-offs.
Nebius pairs GPU clusters with AI Studio managed inference behind an OpenAI-style endpoint where migration usually replaces 1 base_url, hosting mainstream open models with managed fine-tuning and European data centres for EU sovereignty; thin Chinese material and a smaller catalogue are the gaps.
DeepInfra hosts open-weight models on per-token pricing behind an OpenAI-compatible endpoint where migration replaces 1 base_url, covering 4 families in chat, embeddings, reranking and images with serverless elasticity or dedicated instances; the gaps are no closed flagship and cold-start latency.
SambaNova runs inference on in-house RDU silicon through SambaNova Cloud, exposing an OpenAI-style endpoint where migration replaces 1 base_url across mainstream open models with private deployment and long-context support; limited public pricing and high private deployment cost are the gaps.
Hyperbolic combines GPU rental with managed inference behind an OpenAI-compatible endpoint where migration usually replaces 1 base_url, hosting mainstream open models across text and image with hourly or per-token billing; platform scale, hot GPU supply and basic governance are the gaps.