Inkling multimodal model integrated into Hugging Face, vLLM; llama.cpp adds audio input

The latest release of Stockfish 18, a top chess engine, coincides with significant advancements in the open-source AI landscape. Hugging Face Transformers v5.14.0 and vLLM v0.26.0 have integrated the new Inkling multimodal model, a 975 billion parameter model from Thinking Machines, enabling developers to readily use it for various tasks. Concurrently, llama.cpp's b10155 update adds support for MiMo-V2.5 audio input and RVQ-based models, expanding its multimodal capabilities for local inference. AI

IMPACT Expands local inference capabilities for multimodal AI, enabling broader experimentation with new open-weight models.

RANK_REASON Multiple open-source AI model and tool updates, including a new multimodal model and audio input support. [lever_c_demoted from research: ic=1 ai=1.0]

Read on dev.to — LLM tag →

• Hugging Face Transformers
• Inkling
• llama.cpp
• MiMo-V2.5
• NVIDIA
• RVQ
• Stockfish 18
• Thinking Machines
• vLLM

• model release
• product

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →