JetBrains offers Mellum
Product
Mellum is the JetBrains purpose-built coding large language model trained for software developer workflows.
- Currently Offered
Mellum is the JetBrains purpose-built coding large language model trained for software developer workflows.
Mellum is the JetBrains purpose-built coding large language model, trained specifically for code completion, refactoring, and software developer workflows across major programming languages. JetBrains released Mellum in April 2025 and made the base model available as open source on Hugging Face for research and adaptation use.
Mellum powers code completion and AI features inside the JetBrains AI Assistant and Junie agent, complementing third-party models from OpenAI, Anthropic, and Google. JetBrains continues to iterate on Mellum to improve completion quality, context handling, and language coverage for the JetBrains IDE catalog.
Cogito is Deep Cogito's family of open-weight language models with an optional reasoning mode.
Microsoft-Decision-1 is the decision model from Microsoft, delivered through Foundry, that answers bounded questions rather than generating text.
Mistral Small Language Models are open-weight and commercial enterprise models spanning Mistral 7B, Mixtral 8x7B, and Mistral Small.
GLM Model API is Z.ai's developer platform for calling its GLM large language models and agent services over an API.
Open-weights frontier language model family with a 1M-token context window and native text, image, and audio modalities.
Foundation models and AI agents for spatial-temporal reasoning in virtual and physical environments.
Reflection AI builds frontier open-weight language models and autonomous coding agents using large-scale reinforcement learning infrastructure.
AI-powered code review platform that automates pull request reviews to cut review time and bugs in half.
Zencoder IDE Agents embed AI coding agents in VS Code and JetBrains to edit, test, and verify code.
Amazon Q Developer is AWS's generative AI assistant for software development.
Sourcegraph Cody is an AI coding assistant connected to Sourcegraph context.
VSCode coding companion that applies team-curated guidelines to code chat and contribution assistance.
Tabnine is an enterprise AI coding platform with code completion, chat, agents, and an Enterprise Context Engine.
Gemini Code Assist is Google's AI coding assistant for development teams, offered in Standard and Enterprise editions.
Cogito is Deep Cogito's family of open-weight language models with an optional reasoning mode.
Microsoft-Decision-1 is the decision model from Microsoft, delivered through Foundry, that answers bounded questions rather than generating text.
Mistral Small Language Models are open-weight and commercial enterprise models spanning Mistral 7B, Mixtral 8x7B, and Mistral Small.
GLM Model API is Z.ai's developer platform for calling its GLM large language models and agent services over an API.
Open-weights frontier language model family with a 1M-token context window and native text, image, and audio modalities.
Foundation models and AI agents for spatial-temporal reasoning in virtual and physical environments.
DeepSeek-V4 is a frontier open-weight language model developed by DeepSeek.
Reflection AI builds frontier open-weight language models and autonomous coding agents using large-scale reinforcement learning infrastructure.
Cloud AI agent that handles operational engineering work in Slack.
AI-powered code review platform that automates pull request reviews to cut review time and bugs in half.
Zencoder IDE Agents embed AI coding agents in VS Code and JetBrains to edit, test, and verify code.
Z.ai's desktop agentic coding application for macOS, Windows, and Linux, built around its GLM models.
Amazon Q Developer is AWS's generative AI assistant for software development.
Sourcegraph Cody is an AI coding assistant connected to Sourcegraph context.
VSCode coding companion that applies team-curated guidelines to code chat and contribution assistance.
Tabnine is an enterprise AI coding platform with code completion, chat, agents, and an Enterprise Context Engine.
Gemini Code Assist is Google's AI coding assistant for development teams, offered in Standard and Enterprise editions.
Omni is an agentic forward-deployed engineer that builds, runs, and optimizes enterprise AI agents.
