Mythic Inc. (@mythicinc) 's Twitter Profile
Mythic Inc.

@mythicinc

Moving edge AI to a new level with a revolutionary analog compute-in-memory technology.

ID: 869998668153094144

linkhttp://www.mythic-ai.com calendar_today31-05-2017 19:27:01

560 Tweet

2,2K Followers

98 Following

Mythic Inc. (@mythicinc) 's Twitter Profile Photo

Ever wondered how computing at the edge is making an impact? Forbes shares five applications of this approach in practice now, including greater quality control in smart manufacturing and improved decision making for intelligent applications. bit.ly/3HEGbth

Mythic Inc. (@mythicinc) 's Twitter Profile Photo

Curious how Mythic uses analog computing in high performance AI processors to power a breadth of IoT applications more efficiently? In an episode of IoT ONE’s Industrial IoT Spotlight, our CEO Dave Fick explains how this design achievement works. bit.ly/3OZGAem

Mythic Inc. (@mythicinc) 's Twitter Profile Photo

Power optimization is crucial for unlocking the capabilities of today’s technology. Mythic’s Dave Fick shared that analog technologies are key for energy-efficient computing due to its exceptional information density. Head to Semiconductor Engineering for more: bit.ly/3I1AtSE

Mythic Inc. (@mythicinc) 's Twitter Profile Photo

This video showcases Mythic’s use of innovative analog CIM for enhanced AI inferencing in security applications like drones, based on the VisDrone dataset. Mythic’s AMP achieves footage in 1408x1408 resolution with accurate and rapid object detection. bit.ly/3SRKYN4

Mythic Inc. (@mythicinc) 's Twitter Profile Photo

Here’s an example of Mythic’s AMP achieving accurate and rapid AI inferencing results through analog computing. With Mythic, security drones achieve 1408x1408 resolution and high data throughput, increasing reliability in the systems that keep us safe. bit.ly/3SRKYN4

Mythic Inc. (@mythicinc) 's Twitter Profile Photo

This year’s GSMA MWC 2024 unveiled new and exciting directions for AI developments in the IoT. Most notably, AI’s growing impact on connected living and enterprise technologies, including automotive, industrial and smart home. Forbes outlines more here: bit.ly/4c4yKtp

Mythic Inc. (@mythicinc) 's Twitter Profile Photo

With the use of analog computing for AI inferencing, form factors no longer remain a hurdle for IoT devices requiring high performance. Mythic’s Dave Fick shared with SemiWiki of SemiWiki how Mythic’s solutions foster energy efficiency and low latency. bit.ly/3sIWulb

Mythic Inc. (@mythicinc) 's Twitter Profile Photo

As AI’s electricity use raises questions about its economic viability and environmental impact, Bloomberg shares an interesting perspective on how in-memory computing can help make AI more energy efficient. bit.ly/49pvG8y

Mythic Inc. (@mythicinc) 's Twitter Profile Photo

According to Matt Ferrell, the future of AI and computers is analog. We agree. This video from Matt looks at how chipsets, like Mythic’s, conserve energy and enable generative AI products that are more financially feasible and environmentally friendly bit.ly/3xFuvVl

Mythic Inc. (@mythicinc) 's Twitter Profile Photo

Check out the latest episode of Still To Be Decided Podcast from Matt Ferrell, which details how companies like Mythic reinvent analog computing to power the applications of the future. Mythic’s chip can even be tailored to enable hybrid computing. Listen here for more: bit.ly/49MqTye

Mythic Inc. (@mythicinc) 's Twitter Profile Photo

Curious about the future of smart homes? MakeUseOf says to expect AI integration, improved security, and intelligent devices: bit.ly/3UAc5hR. But in order to have viable vision-based solutions, designers need affordable and lowly latent SoCs for quality inferencing.

Mythic Inc. (@mythicinc) 's Twitter Profile Photo

While large language models (LLMs) show great promise, they can present memory bandwidth challenges. If not addressed, their power remains untapped. With Mythic’s in-memory computing, data is pinpointed exactly when it’s needed without a memory cache. bit.ly/47RRk5w

Mythic Inc. (@mythicinc) 's Twitter Profile Photo

.OpenAI’s Chat GPT-4o and Google DeepMind's Project Astra are exciting for the AI industry. But many applications require solutions with enough bandwidth to run LLMs efficiently. Using in-memory computing accelerates the adoption of LLMs across devices. bit.ly/47RRk5w

Mythic Inc. (@mythicinc) 's Twitter Profile Photo

For future-proofed AI-powered products, scalable solutions are needed. Mythic’s AMP offers powerful compute in small edge devices and can scale to serve data centers, providing ultra HD inferencing with support for multiple neural network applications. bit.ly/3pecYzN

Mythic Inc. (@mythicinc) 's Twitter Profile Photo

Dave Fick recently joined MarketScale to discuss the crucial role edge computing plays in the advancement of Smart Cities, noting how local computing can help reduce costs and latencies, scaling applications like cameras and surveillance. bit.ly/3KoE1iM

Mythic Inc. (@mythicinc) 's Twitter Profile Photo

AI and IoT may be new, but the computing fueling the two has a history and distinctly different use cases. Dave Fick shares the unique nature & re-emerging need for analog technology, explaining how it's optimal for AI processing needs on Electronic Design: bit.ly/4bamtlK

Mythic Inc. (@mythicinc) 's Twitter Profile Photo

We’re thrilled to welcome Dr. Taner Ozcelik as our new CEO. Previously building billion-dollar edge computing businesses at Nvidia, his extensive experience and leadership will drive Mythic’s transformative analog computing technology to new heights. bit.ly/4cEBDAF

Mythic Inc. (@mythicinc) 's Twitter Profile Photo

Our new CEO Taner Ozcelik was recognized by Austin Business Journal as a “new name to know” in Austin's technology and startup ecosystem. With Taner’s vision and experience, Mythic will lead the next era of analog-based solutions for high-performance AI applications. bit.ly/3VR4MBO

Mythic Inc. (@mythicinc) 's Twitter Profile Photo

AI technologies consume about 7% of the world’s electricity. According to @EETimes designers need alternative hardware solutions, like Mythic’s, to meet these demands sustainably by reducing electricity use: bit.ly/3S1JsbK

Mythic Inc. (@mythicinc) 's Twitter Profile Photo

Mythic avoids memory bottlenecks during high-resolution inference by storing #AI parameters directly in the processor, a crucial technique for smart security or intelligent manufacturing, which requires energy efficiency and timely processing. Learn more: bit.ly/3LH6EZg