Docker

Kimi K2 Thinking: open-source agent with deep reasoning, stable tool use, fast INT4, 256k context.

Kimi K2 Thinking: open-source agent with deep reasoning, stable tool use, fast INT4, 256k context.

DeepSeek-V3.2 boosts efficiency and reasoning with DSA, scalable RL, agentic data—IMO/IOI wins.

Ministral 3: compact vision-enabled model with near-24B performance, optimized for local edge use

Ministral 3: compact vision-enabled model with near-24B performance, optimized for local edge use

Multilingual reranking model for text retrieval, scoring document relevance across 119 languages.

Multilingual reranking model for text retrieval, scoring document relevance across 119 languages.

Snowflake’s Arctic-Embed v2.0 boosts multilingual retrieval and efficiency

Qwen3 Embedding: multilingual models for advanced text/ranking tasks like retrieval & clustering.

Qwen3 Embedding: multilingual models for advanced text/ranking tasks like retrieval & clustering.

OpenAI’s open-weight models designed for powerful reasoning, agentic tasks

The most advanced Qwen model yet, with major gains in text, vision, video, and reasoning.

Safety reasoning models for policy-based text classification and foundational safety tasks.

Qwen3 is the latest Qwen LLM, built for top-tier coding, math, reasoning, and language tasks.

Granite-4.0-nano: lightweight instruct model trained via SFT, RL, and merging on diverse data.

Granite-4.0-h-nano: lightweight instruct model trained via SFT, RL, and merging on diverse data.

Google’s latest Gemma, small yet strong for chat and generation

OpenAI’s open-weight models designed for powerful reasoning, agentic tasks

Embedding Gemma is a state-of-the-art text embedding model from Google DeepMind

all-MiniLM-L6-v2 is a sentence-transformers, maps sentences & paragraphs to a 384 dimensional vector

Qwen3 is the latest Qwen LLM, built for top-tier coding, math, reasoning, and language tasks.

SmolVLM: lightweight multimodal model for video, image, and text analysis, optimized for devices

Granite Docling is a multimodal model for efficient document conversion.

An open-source visual language model that interprets images via text prompts, fast and powerful.

SmolVLM: lightweight multimodal model for video, image, and text analysis, optimized for devices.

32B long-context instruct model with RL alignment, IF, tool use, and enterprise optimization.

7B long-context instruct model with RL alignment, IF, tool use, and enterprise optimization.

3B long-context instruct model with RL alignment, IF, tool calling, and enterprise readiness.

3B long-context instruct model with RL alignment, IF, tool use, and enterprise optimization.

Agentic coding LLM (24B) fine-tuned from Mistral-Small-3.1 with a 128K context window