Jake Cunningham

Machine Learning @ UCL

[Artificial Analysis @ArtificialAnlys]

Cartesia’s Sonic-3.5 takes the #1 spot on the Artificial Analysis Speech Arena Leaderboard, surpassing Inworld Realtime TTS 1.5 Max and Google’s Gemini 3.1 Flash TTS

Sonic-3.5 is the latest TTS model from @cartesia . It supports 42 languages, including 9 Indian languages, with 500+ voices available out of the box. The model has been highly preferred among voters in the TTS Arena, with its demonstrated naturalness and accurate transcript following.

Key takeaways:
➤ Quality: Sonic-3.5 has an Elo score of 1,218 (+16/-16) based on 1,144 arena appearances, placing it ahead of Inworld Realtime TTS 1.5 Max at 1,194 and Gemini 3.1 Flash TTS at 1,209

➤ Pricing: Sonic-3.5 is priced at $39/1M characters, a premium compared to Gemini 3.1 Flash TTS at $18.3/1M characters, and Inworld Realtime TTS 1.5 Max at $35/1M characters

➤ Speed: 105.5 characters per second, compared to 205 characters per second for Inworld Realtime TTS 1.5 Max and 26.3 characters per second for Gemini 3.1 Flash TTS

See more details and listen to samples below 🧵

[Jean-Gabriel BARTHELEMY @JG_Barthelemy]

We’ve been lucky enough to test Mamba-3 ahead of the curve. 🧪

Here is how it integrates into Hybrid Models (Spoiler: it unlocks Muon for SSMs for the first time). 🧵

[Cartesia @cartesia]

Mamba-3 is out! 🐍

SSMs marked a major advance for the efficiency of modern LLMs.

Mamba-3 takes the next step, shaping SSMs for a world where AI workloads are increasingly dominated by inference.

Read about it on the Cartesia blog:

https://t.co/dIWg3iXfay

[Albert Gu @_albertgu]

The newest model in the Mamba series is finally here 🐍

Hybrid models have become increasingly popular, raising the importance of designing the next generation of linear models.

We've introduced several SSM-centric ideas to significantly increase Mamba-2's modeling capabilities without compromising on speed. The resulting Mamba-3 model has noticeable performance gains over the most popular previous linear models (such as Mamba-2 and Gated DeltaNet) at all sizes.

This is the first Mamba that was student led: all credit to @aakash_lahoti @kevinyli_ @_berlinchen @caitWW9, and of course @tri_dao!

[Albert Gu @_albertgu]

we're making a focused push on model innovations this year @cartesia! hiring actively, including in the London office

[David W. Romero @davidwromero]

HIRING!

We’re looking for strong PhD students to join our core research team @cartesia — working w/ @_albertgu, @HJCunningham97, @MucaCirone, @DanielePaliotta, me & more

Topics: multimodal, long-context & tokenization-free / -dynamic methods

SF & London

Interested? DM me!

[Martin Jørgensen @JorgensenMart]

💣I am excited to advertise 'Bézier Gaussian Processes for Tall and Wide Data' which is to appear at #NeurIPS2022

The paper presents a novel approach to scalable GP regression through a structured control points, which can be seen as a type of 'inducing points'. More in thread.

[Milan Klöwer @milankloewer]

Today's magic trick: Make vorticity look like clouds ☁️🌧️ to celebrate that we just released SpeedyWeather.jl v0.3, our still-baby weather model in #JuliaLang. #GPU, auto-differentiation, low-precision, machine learning, all coming soon! Check it out, https://t.co/h9k6FiOUF4

[Journal of Machine Learning Research @JmlrOrg]

JMLR has recently launched a Special Issue on ML for addressing problems of climate change! We welcome all submissions which use machine learning to address problems of climate change, including mitigation, adaptation, and climate science. [1/4]

[Ton van den Bremer @tonvandenbremer]

A new paper on how the Coriolis force can change the way surface waves transport marine plastic: https://t.co/PBzSRHB9TN in #jgroceans @theAGU funded by @RAEngNews!