Preferred Networks

Preferred Networks, Inc. (PFN)'s official account
Japanese account: @PreferredNetJP

[Preferred Networks @PreferredNet]

PFN and Tenstorrent's joint R&D proposal has been selected for a NEDO-sponsored project. Our goal is to use photonics-electronics integration and build an AI processor system that delivers high performance and efficiency for both training and inference.
https://t.co/5SGT4hY67F

[Preferred Networks @PreferredNet]

📢PFN is accepting pre-applications for 2028 new grads
Registrants will receive updates regarding our official application process and tech events for students.
Pre-apply here: https://t.co/NcXCRClH6w

[Preferred Networks @PreferredNet]

PFN group delivering full-stack AI tech from custom chips to end-user service: PFN and PFCI will start accelerating selected calculations (for crystal structures, etc.) on the Matlantis™ materials simulator with PFN's MN-Core™ 2 AI chips in 2026.
https://t.co/8iwewL9m5u

[Preferred Networks @PreferredNet]

Mitsubishi Heavy Industries and PFN have entered into an agreement under which MHI plans to invest 10 billion yen in PFN to jointly develop AI technologies to enhance the intelligence and autonomy of machinery and systems across mission-critical domains. https://t.co/rLAajClEok

[OptunaAutoML @OptunaAutoML]

Today we're launching “Rustuna”, a faster Optuna implementation written in Rust ⚡️

Rustuna is designed as a faster alternative to Optuna, with a familiar API, a lower memory footprint, and no Python runtime dependencies. See how it compares with Optuna in the benchmark below 👇

[Preferred Networks @PreferredNet]

During Steam's official event PvE Survival Crafting Fest 2026, we will hold a 40% discount sale to celebrate our PC game “Omega Crafter” surpassing 100,000 players (including demo) https://t.co/vxKL2Qhd5d

[Preferred Networks @PreferredNet]

Together with the Qualcomm Japan Robotics Center, we will leverage PFN's foundation model expertise, Qualcomm' computing platforms and Japan's manufacturing ecosystem to accelerate deployment of practical physical AI applications in Japan. https://t.co/PKWzGw9go1

[Preferred Networks @PreferredNet]

How does Matlantis-PFP v9 perform on each MLIP Arena task? We published task-by-task results, covering five tasks: homonuclear diatomics, equation of state, energy-volume scans, stability, and combustion. https://t.co/35kSihuNqt

[Preferred Networks @PreferredNet]

We released Matlantis-PFP v9. Elemental coverage now spans 96 elements (H–Cm) including lanthanides and actinides, with improved experimental agreement via r2SCAN. We also discuss the benchmark evaluation using MLIP Arena and future development directions.
https://t.co/2xQ36jOiYa

[Preferred Networks @PreferredNet]

Please exercise caution: We have recently confirmed suspicious communications in which unrelated third parties are conducting recruitment activities while falsely claiming to represent our company or our group companies, often using fictitious job titles.
https://t.co/p6NXgyaTwL

[Preferred Networks @PreferredNet]

PFN 2025 summer intern Boming Yang explores the translation and evaluation challenges in building a strong PLaMo™-based Japanese reward model. https://t.co/s61kXfo2r3

[Preferred Networks @PreferredNet]

Preferred Networks will invest in Noetra and participate in its project focused on Japan-developed multimodal foundation model for Physical AI. https://t.co/hBEAYsOTyr

[Preferred Networks @PreferredNet]

At a recent 🇫🇷CEA-Leti event, Takahiro Ogura, President of PFN’s AI Computing Division, introduced the MN-Core™ Partner Program. We’re inviting partners to help build businesses powered by MN-Core AI chips! Details👉 https://t.co/Zo3lvZP9lL

[Preferred Networks @PreferredNet]

PFN CFO Yotaro Katayama joined the Japan-India Joint Economic Forum in Delhi yesterday in support of the 🇮🇳🇯🇵 PMs' joint statement on AI cooperation. Looking forward to future collaborations with Indian companies!

[Preferred Networks @PreferredNet]

PFN has teamed up with Mitsubishi Heavy Industries (MHI) to make Japan-made AI for mission-critical fields by combining MHI’s hardware, control & simulation technologies with PFN’s AI expertise. We also aim to establish a capital alliance within FY2026. https://t.co/rDngxDOM0O

[トヨタ 未来創生センター | Toyota Frontier Research Center @TOYOTA_FRC]

#FrontierResearchCenter and Preferred Networks, Inc. are conducting joint research on physical AI that leverages large-scale data. Today, PFN announced a collaborative research initiative to accelerate inference processing by integrating PFN’s MN‑Core™ L series—an inference-

[Preferred Networks @PreferredNet]

We have started joint research with Toyota's Frontier Research Center to accelerate physical AI using our upcoming ultra-high-bandwidth AI inference chips in the MN-Core L Series for robots which typically require high-speed on-premise inference
https://t.co/iIeD7RhljA

[Preferred Networks @PreferredNet]

PFN Co-Founder & CEO @hillbig will speak at Imec Technology Forum #ITFWorld2026 in Antwerp, Belgium at 4:30pm CET today. Theme: "Chips for AI, AI for Chips: Redesigning AI from Materials up" Abstract: https://t.co/X5fA8kJwWZ

[Preferred Networks @PreferredNet]

Visit the PFN desk in the Microsoft Japan booth (A103, West 3) at #SusHiTechTokyo2026 today and tomorrow! PFN's global biz VP and MS Japan's CTO will be on Green Stage (1st floor) today at 17:15-18:00 to discuss integration of Azure between our PLaMo LLM
https://t.co/a4mhhbtdg5

[Preferred Networks @PreferredNet]

Hiring now: MN-Core runtime software engineer https://t.co/vuAfKFd1Sr

[Preferred Networks @PreferredNet]

Hiring now: We started looking for engineers who can develop the LLM serving engine with us for our MN-Core™ L series of proprietary inference AI chip.
See reply for other MN-Core-related positions👇 https://t.co/heSbL4cm7n

[Preferred Networks @PreferredNet]

Hiring now: MN-Core compiler engineer
https://t.co/VwBNYU5nqj