Meta releases Muse Spark 1.3 with frontier-level performance, stronger coding and agentic abilities, and aggressive pricing; early benchmarks put it at #3 betwe…
Meta’s Muse Spark 1.3 reportedly takes first place on DeepSWE with 75.4%, ahead of GPT-5.6 Sol and Fable 5. The model is rolling out in Muse Code and the Meta M…
Stability AI shares key takeaway from Montreal hackathon: the future of creative AI lies in smaller, locally running models that feel intimate and integrated in…
GMI Cloud is giving users five more days of free access to MiniMax M3, M2.7, Music 3.0, and Speech 2.8. It also invites creators to join the MiniMaxathon for a…
Meta has released Muse Spark 1.3, its fourth model release in five months. The new version shows strong benchmark gains, especially in agentic and scientific ca…
Grok Imagine now supports up to 14 references (images, voices, character references) per video prompt, using @ tags.
Command Code adds Gemini 3.8 Flash to its GOAT offering, giving users 4x usage of the model for AI-assisted coding.
Muse Spark 1.3 is rolling out today with frontier-level performance in coding and agentic work at very low cost, available via Muse Code and the API. Ollama is…
Perplexity has made Lily, its local inference engine built for Apple silicon, available. Lily treats Apple silicon as a distinct inference platform and maps Qwe…
Perplexity shared benchmark results for its open-source local inference engine Lily on an M5 Max MacBook Pro, showing 1.23× higher prefill throughput and 1.35×…
