Model / Foundation

General-purpose large language models, including open-weight model families.

Ordered by attributes answered — the column we hold for most of them. Blank where we hold none.

AI21 Labs6 ATTRIBUTES ANSWEREDAllenAI OLMo6 ATTRIBUTES ANSWEREDSnowflake Arctic6 ATTRIBUTES ANSWEREDAnthropic5 ATTRIBUTES ANSWERED01.AI4 ATTRIBUTES ANSWEREDCohere4 ATTRIBUTES ANSWEREDIBM Granite4 ATTRIBUTES ANSWEREDMeta Llama4 ATTRIBUTES ANSWEREDOpenAI4 ATTRIBUTES ANSWEREDDeepSeek3 ATTRIBUTES ANSWEREDfunclip3 ATTRIBUTES ANSWEREDGoogle DeepMind3 ATTRIBUTES ANSWERED

Models whose output is an image, a video or a 3D asset: diffusion and transformer weights, and the libraries that serve them closely. A model that reads several modalities and answers in text is multimodal; a finished tool somebody opens to make something is a creative application.

Ordered by attributes answered — the column we hold for most of them. Blank where we hold none.

minimax-mcp3 ATTRIBUTES ANSWEREDods3 ATTRIBUTES ANSWEREDAlibaba Qwen2 ATTRIBUTES ANSWEREDOOpenMontage Studio2 ATTRIBUTES ANSWEREDCclip-guided-diffusion1 ATTRIBUTE ANSWEREDDdiffsinger1 ATTRIBUTE ANSWEREDGgpt-image2-skill1 ATTRIBUTE ANSWEREDlakonlab1 ATTRIBUTE ANSWEREDLightricks1 ATTRIBUTE ANSWEREDLuma AI1 ATTRIBUTE ANSWEREDopen-generative-ai1 ATTRIBUTE ANSWEREDpromptenhancer1 ATTRIBUTE ANSWERED

Models that take or produce more than text: images, video, audio in one context window. Text-only models are large language models; single-modality speech belongs in speech and audio.

Ordered by attributes answered — the column we hold for most of them. Blank where we hold none.

Mistral AI9 ATTRIBUTES ANSWEREDopenrlhf4 ATTRIBUTES ANSWEREDcamel3 ATTRIBUTES ANSWEREDdia3 ATTRIBUTES ANSWEREDPparlor3 ATTRIBUTES ANSWEREDReka AI2 ATTRIBUTES ANSWEREDZhipu AI2 ATTRIBUTES ANSWEREDDdinov3-yolov12NO ANSWER ON RECORDgroundinglmmNO ANSWER ON RECORDUltralyticsNO ANSWER ON RECORD

Models whose output is a vector rather than a reply — text, image and code embeddings, and rerankers. Where those vectors are stored is a vector database.

Ordered by attributes answered — the column we hold for most of them. Blank where we hold none.

kubeai4 ATTRIBUTES ANSWEREDBbcembedding2 ATTRIBUTES ANSWEREDcolivara2 ATTRIBUTES ANSWEREDcleora1 ATTRIBUTE ANSWEREDCuriosity1 ATTRIBUTE ANSWEREDGgraphrag-local-ollama1 ATTRIBUTE ANSWEREDai-dial-coreNO ANSWER ON RECORDMinishNO ANSWER ON RECORD

Domain-specific models for one industry such as medicine, law or biology.

Ordered by attributes answered — the column we hold for most of them. Blank where we hold none.

SQLRite3 ATTRIBUTES ANSWEREDStability AI3 ATTRIBUTES ANSWEREDcvxpy1 ATTRIBUTE ANSWEREDacadosNO ANSWER ON RECORDfingptNO ANSWER ON RECORDYouTrackDBNO ANSWER ON RECORD