Multiple new models are now generally available in GitHub Copilot 🚀

[Akash1134]

Anthropic Claude 3.7 Sonnet, Claude 3.5 Sonnet, OpenAI o3-mini, and Google Gemini Flash 2.0 have leveled up and are now generally available in GitHub Copilot. With this update, these models move from our preview release terms to generally available release terms, extending indemnification for IP infringement to code generated with these models in Copilot Chat and agent mode.

• Claude 3.7 Sonnet: Anthropic's most advanced model yet, it excels in development tasks that demand structured reasoning across large or complex codebases.
• Claude 3.5 Sonnet: A reliable companion for everyday coding support.
• OpenAI o3-mini: A speedy and cost-effective reasoning model designed to deliver top-notch coding performance with low latency and resource usage.
• Gemini Flash 2.0: Google’s model, optimized for rapid responses and dynamic multimodal interactions.

Discover more about the models available in Copilot by checking out our documentation and get started with Copilot today.

Got questions or feedback? Share them in the comments. 🚀

[D4n16l6]

A parer mio il modello più potente è Sonnet 3.7 Thinking da quello che ho potuto vedere. Specialmente nella risoluzione dei problemi di logica e nella individuazione e proposta di soluzioni relative ai bug. Gli altri modelli sono in genere più veloci ma soffrono di bias cognitivi, GEMINI 2.0, per esempio, tende a inventare variabili e funzioni dicendo che le vede nel codice mentre è falso e solo dopo molta insistenza si riesce ad ottenere una risposta più o meno valida, ovviamente quest non è aiuto ma complicazione.
In poche parole un valido aiuto nella programmazione potrebbe venire da modelli che riescono a scorrere tutto il progetto (circa 7000 righe) e che "pensino" attentamente alla soluzione da proporre.

[kkc-tonywu]

Hi @Akash1134,

I am the admin of the organization, and I was very excited to see the announcement that multiple new models are now generally available in GitHub Copilot GA. As a result, I immediately enabled Claude 3.7 Sonnet, Claude 3.5 Sonnet, OpenAI o3-mini, and Gemini Flash 2.0.

However, in the Copilot Chat on VS Code, I only see GPT4o and Claude 3.5 Sonnet available for selection.

Is this the expected behavior?

Version Information:

• VSCode: Version 1.99.0
• GitHub Copilot: 1.296.0
• GitHub Copilot Chat: 0.26.0

[kkc-tonywu]

I checked after a day, and the models are now displayed. What could be the reason for this situation?

[Ammber666]

Some time ago, my student certification was approved. However, I am unable to use Claude 3.7 and O1.

[Akash1134]

I checked after a day, and the models are now displayed. What could be the reason for this situation?

Hi @kkc-tonywu , The deployment of models update is done in a phased manner that might've caused the situation. Hope thah helps!

[Ammber666]

Some time ago, my student certification was approved. However, I am unable to use Claude 3.7 and O1.

[yuqianten]

怎麼回事

[Akash1134]

Hi @Ammber666 Please note that Clicking Allow enables you to use Claude Sonnet and updates the policy in your personal settings on GitHub.

You can enable the model directly in your personal settings on the GitHub website. See Managing Copilot policies as an individual subscriber. Let us know how it goes.

[Christian-Vector]

The Claude 3.7 and 3.7 Thinking Model is causing an error in the Eclipse Copilot Chat:

java.lang.IllegalStateException: Turn step failed, title=Generating response
	at com.microsoft.copilot.eclipse.ui.chat.ChatView.onChatProgress(ChatView.java:256)
	at com.microsoft.copilot.eclipse.core.chat.ChatProvider.notifyProgress(ChatProvider.java:43)
	at com.microsoft.copilot.eclipse.core.lsp.CopilotLanguageClient.notifyProgress(CopilotLanguageClient.java:90)
	at java.base/jdk.internal.reflect.DirectMethodHandleAccessor.invoke(DirectMethodHandleAccessor.java:103)
	at java.base/java.lang.reflect.Method.invoke(Method.java:580)
	at org.eclipse.lsp4j.jsonrpc.services.GenericEndpoint.lambda$recursiveFindRpcMethods$0(GenericEndpoint.java:65)
	at org.eclipse.lsp4j.jsonrpc.services.GenericEndpoint.notify(GenericEndpoint.java:160)
	at org.eclipse.lsp4j.jsonrpc.RemoteEndpoint.handleNotification(RemoteEndpoint.java:231)
	at org.eclipse.lsp4j.jsonrpc.RemoteEndpoint.consume(RemoteEndpoint.java:198)
	at org.eclipse.lsp4e.LanguageServerWrapper.lambda$3(LanguageServerWrapper.java:356)
	at org.eclipse.lsp4j.jsonrpc.json.StreamMessageProducer.handleMessage(StreamMessageProducer.java:185)
	at org.eclipse.lsp4j.jsonrpc.json.StreamMessageProducer.listen(StreamMessageProducer.java:97)
	at org.eclipse.lsp4j.jsonrpc.json.ConcurrentMessageProcessor.run(ConcurrentMessageProcessor.java:114)
	at java.base/java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:572)
	at java.base/java.util.concurrent.FutureTask.run(FutureTask.java:317)
	at java.base/java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1144)
	at java.base/java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:642)
	at java.base/java.lang.Thread.run(Thread.java:1583)

Is that a known Problem?

[Akash1134]

Hi @Christian-Vector,

Based on the error log you shared, there appears to be a compatibility issue between these newer Claude models and the Eclipse Copilot integration, resulting in the IllegalStateException when generating responses. Please ensure that your GitHub Copilot for Eclipse extension updated, as any compatibility fixes would be released through regular updates.

Thanks,
Akash

[yuqianten]

Nice

[iwinux]

The price of o3 has been lowered. Time to update the premium request multiplier :)