GitHub today announced it will now let developers switch between several large language models when they are using Copilot Chat, the company's code-centric service similar to ChatGPT. Until now, Copilot Chat was powered by OpenAI's GPT-4. Going forward, developers can choose from Anthropic's Claude 3.5 Sonnet, Google's Gemini 1.5 Pro, or OpenAI's GPT-4o, o1-preview, and o1-mini. Anthropic will be first out of the gate, followed a few weeks later by Gemini.

Some pundits might view this as another way for Microsoft to cut its dependence on OpenAI, but GitHub CEO Thomas Dohmke framed it in terms of giving developers a choice. Not every model, after all, excels at every development-related task, and some models are simply better at working with certain languages than others.

We truly believe that the era of a single model is over," Dohmke told me. "As you know, GitHub Copilot itself has always used multiple versions of the GPT-3 or -4 models for different tasks. Where low latency was required, we were using a faster, smaller model. Where we needed higher accuracy, we were using a bigger, more complex model.". We believe in developer choice," the GitHub spokesperson said, and developers-owing to company policy, the benchmarks that they've seen, different programming languages and, of course, personal preference, or because they're using that model for other scenarios already-prefer one of the competing models, and so we're officially partnering with both Anthropic and Google."

The team had done a few pieces of prep work for this launch because when it started offering developers a choice between GPT-4 and GPT-4 o1, the latter just launched a bit more than a month ago.
This choice so far only applies to Copilot Chat and the new tool that has just been released called Spark, although the company would like to add this choice to all its tools, Dohmke noted.

A vast choice of models, from code generation, refactoring, and optimization to anything else developers need to optimize development, is something they have been asking for, Google Cloud CEO Thomas Kurian said. Gemini models perform exceptionally well in that aspect and are now accessible to millions of developers all around the world on popular platforms and environments — and even GitHub Copilot now, as well — to gain enterprise-grade trusted AI using Google Cloud.