Models

Open vision-language MoE model for image and video understanding, documents, and visual agents

Open-weight hybrid model for enterprise chat, coding, retrieval-augmented generation, and tool-calling workloads

Compact open-weight hybrid Granite model for lightweight enterprise chat and tool calling

Late GLM-4 workhorse for coding agents, reasoning, and structured tasks

Open-weight instruction model for adaptable chat and self-hosted production workloads

Gemma 3 27B tuned by AI Singapore for Southeast Asian languages and instruction following

Qwen vision-language thinking model for visual reasoning, documents, and agent tasks

Qwen vision-language instruct model for visual reasoning, documents, and agent tasks

Multimodal reasoning model accepting text, images, video, and audio with text output

Open multimodal reasoning model for transparent analysis of text and images

September 2025 Kimi K2 instruction update with a 256K context and improved agentic coding

Fully open 70B multilingual LLM supporting 1800+ languages with 65K context. Trained on 15T tokens of compliant open data. Apache 2.0, EU AI Act compliant.

Fully open 8B multilingual LLM supporting 1800+ languages with 65K context. Trained on compliant open data. Apache 2.0, EU AI Act compliant.

Qwen instruction model for multilingual chat, reasoning, and tool use

Efficient Qwen thinking model for local reasoning, math, and coding agents

Reasoning model for deliberate analysis, multi-step problem solving, and tool use

DeepSeek chat model for instruction following, coding, and analysis

Cohere reasoning model for multilingual enterprise agents, tools, and complex workflows

Hybrid-reasoning DeepSeek model with thinking and non-thinking modes

Open ByteDance Seed model for long-context reasoning, tool use, and agents

Compact Nemotron model for efficient reasoning and deployable AI agents

Open medical model trained for clinical reasoning and patient interaction with on, off, and auto thinking modes

GLM vision model for visual reasoning, documents, and multimodal agents

Open GPT reasoning model for self-hosted agents and controllable deployments

Open GPT reasoning model for self-hosted agents and controllable deployments

Smaller Qwen coder for efficient local agents and repo-level fixes

Cohere vision model for multilingual document analysis, OCR, and image understanding

Hybrid-reasoning GLM release that made the 4.5 line broadly useful

Lighter GLM-4.5 variant for fast coding assistance and cheaper agents

Updated compact Qwen3 MoE instruct model for multilingual chat, coding, and tool use

Nemotron model for efficient reasoning, coding, and specialized AI agents

Updated large open Qwen3 MoE thinking model for reasoning, math, coding, and tool use

Open Qwen coding heavyweight for repository reasoning and agentic engineering

Updated large open Qwen3 MoE instruct model for multilingual chat, coding, and tool use

Instruct model with native audio input for speech understanding and tool use

Open audio-language model for speech transcription, audio understanding, and voice-driven tool use

Open audio-language model for speech transcription, audio understanding, and voice-driven tool use

Kimi model for long-context chat, coding, and agentic reasoning

Non-thinking Kimi K2 MoE instruction model for coding and tool-using agents

Mistral coding agent model for repository tasks and software engineering workflows

Mistral model for multilingual chat, reasoning, and tool-assisted workflows

AI21's hybrid SSM-Transformer long-context model for enterprise agents and grounded generation

Multimodal reasoning model for visual analysis, planning, and tool use

Open ERNIE text MoE with 21B total and 3B active parameters

Open ERNIE text MoE with 300B total and 47B active parameters

Efficient on-device Gemma 3n model for text, image, and audio understanding

Efficient on-device Gemma 3n model for text, image, and audio understanding

Efficient Mistral model for fast chat, extraction, and production assistants

Efficient Mistral model for fast chat, extraction, and production assistants

Open hybrid-attention MoE reasoning model with a million-token context and an 80K thinking budget