Blog
Read about our latest announcements.
What you Actually Get (and Don’t Get) with GitHub Copilot for Business
Clarifying the GitHub Copilot for Business offering.
Productionizing Context Aware Everything
Why it is hard to create a context reasoning engine for code LLMs that consistently works.
Context Aware Everything: More Advanced Realtime Context than GitHub Copilot
A deep dive into context awareness of Codeium and how it stacks up against GitHub Copilot and CopilotX.
What to Know About the Context Going into your Code LLM
Why real-time context for AI code assistants is a meaningful and tricky problem.
GitLab Code Suggestions: Why You Should Not Build Your Own Generative AI Product for Code
An analysis on the capabilities and performance of the AI code assistant from GitLab.
What GitHub Copilot Lacks: Fine-tuning on Your Private Code
Proof that Codeium fine-tuned on a repository significantly outperforms GitHub Copilot.
Code Suggestions You Don’t Get from Copilot: In-line FIM
In-line Fill-in-the-Middle suggestions, valuable suggestions produced only by Codeium.
Should your Company ban Generative AI?
Generative AI poses risks for companies that do not have strict data governance
Amazon CodeWhisperer is out of beta. We tried it. Spoiler: it isn’t good.
An in-depth analysis on the capabilities and performance of Amazon CodeWhisperer post-general access release.
GitHub Copilot Emits GPL. Codeium Does Not.
Demonstrating that GitHub Copilot trains on non-permissive licenses and cannot filter suggestions properly, while Codeium avoids legal risk.
User Personalization, the Next Frontier in LLMs
How tuning the model layer of LLM applications creates the highest quality experiences for enterprises.
Why your AI Code Completion tool needs to Fill in the Middle
Analyzing how fill-in-the-middle allows Codeium to make better suggestions.