Ollama’s v0.40.13-rc0 includes LFM3 GGUF fixes, llama-server speech support and an updated llama.cpp version.
NoteBy OossaPublished by Oossa: 1 min read
Ollama’s v0.40.13-rc0 release includes fixes for LFM3 GGUF models and speech support in llama-server. It also updates llama.cpp to version b11963. The release notes do not explain which LFM3 issues were fixed or what speech support enables, so the practical changes for users are not specified.
People running Ollama may get improved LFM3 GGUF handling and llama-server speech support, but the release notes do not describe the specific fixes or capabilities.
Models
Artificial Analysis gives Google’s Gemini 4 Carbon the top score on its Intelligence Index v5. Google is rolling it out to selected users, not the public.
Models
Google is rolling out Chrome autofill suggestions based on Gmail for US users who have Gemini’s Personal Intelligence enabled.
Models
The open-weight model transcribes 40 languages and can translate, label speakers and answer questions about audio. Developers can download it from Hugging Face and ModelScope.
Oossa · Newsletter
The week in AI, explained
Every Monday: the stories worth knowing, in plain language. Free, no spam.
