SWE-1.5
SWE-1.5 is the latest agent-model release by Cognition, purpose-built for software engineering and characterized by a “frontier-size” architecture comprising hundreds of billions of parameters and optimized end-to-end (model, inference engine, and agent harness) for both speed and intelligence. It achieves near-state-of-the-art coding performance and sets a new benchmark in latency, delivering inference speeds up to 950 tokens/second, roughly six times faster than its predecessor Haiku 4.5 and thirteen times faster than Sonnet 4.5. The model was trained using extensive reinforcement learning in realistic coding-agent environments with multi-turn workflows, unit tests, quality rubrics, and browser-based agentic execution; it also benefits from tightly integrated software tooling and high-throughput hardware (including thousands of GB200 NVL72 chips and a custom hypervisor infrastructure).
Learn more
Cody
Cody, Sourcegraph’s AI code assistant goes beyond individual dev productivity, helping enterprises achieve consistency and quality at scale with AI.
Unlike traditional coding assistants, Cody understands the entire codebase, enabling deeper contextual awareness for smarter autocompletions, refactoring, and AI-driven code suggestions. It integrates with IDEs like VS Code, Visual Studio, Eclipse, and JetBrains, providing inline editing and chat without disrupting workflows. Cody also connects with tools like Notion, Linear, and Prometheus to enhance development context. Powered by advanced LLMs like Claude Sonnet 4 and GPT-4o, it optimizes speed and performance based on enterprise needs, and is always adding the latest AI models. Developers report significant efficiency gains, with some saving up to six hours per week and doubling their coding speed.
Learn more
Amp
Amp is a frontier coding agent built to give developers full access to the power of today’s leading AI models directly in their workflow. Available in the terminal and popular editors like VS Code, Cursor, Windsurf, JetBrains, and Neovim, Amp integrates seamlessly into existing development environments. It enables developers to delegate complex coding tasks, refactors, reviews, and explorations to intelligent agents that understand and operate across entire codebases. With support for advanced models such as Claude Opus, Gemini, and GPT-class models, Amp delivers fast, reliable, and highly agentic code generation. The platform is designed for real-world engineering work, handling multi-file changes, deep context, and iterative improvements. Amp helps developers move faster while maintaining confidence in code quality.
Learn more
GitHub Copilot
GitHub Copilot is an AI-powered coding companion that helps developers write, debug, and refactor code faster than ever. Built directly into popular IDEs like VS Code, JetBrains, and Visual Studio, Copilot suggests context-aware code completions and automates repetitive tasks. With Agent Mode, Copilot can plan, write, test, and even submit pull requests—acting like an onboarded team member that understands your repositories. Developers can toggle between advanced models such as GPT-5, Claude Opus 4.1, and Gemini 2.5 Pro to balance speed and depth. Its Copilot Chat and Spaces features bring code, documentation, and project notes together for richer, team-tailored insights. Whether you’re building solo or leading an enterprise project, Copilot enhances productivity and creativity across the entire software lifecycle.
The GitHub Copilot coding agent streamlines development by automating repetitive and time-consuming tasks, allowing developers to focus on more complex coding.
Learn more