v0.0.30

What's Changed

🚀 Main Features

• Implemented CodeGemmy model for enhanced code completion.
• Utilized a local tokenizer for improved performance and more accurate prediction times.

🛠️ Updates and Enhancements

• Update server builds to late llama.cpp(b2665) by @gespispace in #28
• feat(configuration): add configuration option to use pre release by @gespispace in #27
• bugfix(configuration): fix type in default config for cloud use by @gespispace in #29
• refactor(ci): move s3 url for server builds to new firecoder domain by @gespispace in #30
• bugfix(chat): fix references for highlighted text by @gespispace in #31
• feat(completion): move to codegemmy completion by @gespispace in #32
• feat(configuration): auto restart servers after changing configuration by @gespispace in #33
• feat(chat): add cancel button in chat to stop generation by @gespispace in #36
• feat(prompt): use tokenizer from @xenova/transformers by @gespispace in #38
• feat(cloud): auto completion by @gespispace in #39
• feat(chat): add copy button in chat by @gespispace in #41

Full Changelog: v0.0.29...v0.0.30