Tag
Cerebras’ share price has halved in the five months since its IPO (falling from $311 to $109). The bulk of its $25.4 billion in orders comes from OpenAI, and the article examines concerns about its ability to deliver and the competition it faces, including Nvidia’s acquisition of Groq.
Cerebras co-founder and CEO Andrew Feldman explains why the company's wafer-scale architecture claims ~2,500× faster memory-to-compute movement than GPUs during LLM decode, since Cerebras keeps model weights in on-chip SRAM rather than HBM.
Sam Altman addresses speculation about OpenAI's partnership with Cerebras, confirming it as a close partnership with deep engagement focused on pushing the frontiers of speed.
Cerebras Systems CEO Andrew Feldman will headline TechCrunch Disrupt 2026 with a talk titled 'Can AI Keep Scaling?', covering compute demand, energy constraints, and Cerebras' wafer-scale approach to AI hardware.
General Compute is deploying Cerebras systems alongside NVIDIA GPUs to offer 20x faster AI inference, funded by $400M in debt, to address compute availability for developers.
OpenAI is preparing to expand its Ultrafast API mode to more users, with potential broader availability during DevDay, powered by Cerebras for faster inference speeds.
Wafer outperformed Cerebras on latency for YC's AI Office Hours, with GLM-5.2 achieving 379 ms versus 674 ms for Gemma 4, resulting in 44% lower latency and increased user engagement.
Researchers built a faster AI assistant using Qwen 3.8 27B on Cerebras with Pi harness, achieving a 19x speed improvement over existing assistants through optimized parallel checks and faster inference.
Andrew Feldman, CEO of Cerebras, discusses how building Cerebras at scale has given him a new appreciation for Jensen Huang's work at Nvidia, highlighting the grit required in leadership.
Andrew Feldman, CEO of Cerebras, highlights the intense excitement around AI in Silicon Valley, with anecdotes of early morning meetings showcasing the industry's fervor.
OpenAI uses Cerebras for fast inference to improve incident response and critical research during outages, as described by @seanlie.
The article analyzes the shift in AI focus from training to inference in 2026, driven by hardware innovations and strategic alliances among tech giants like Nvidia, Amazon, and Cerebras to meet growing demand.
A demonstration of an operating system powered by Qwen 3.8 27B running at 1950 tokens per second on Cerebras hardware, where the model weights act as the runtime for real-time software generation.
A social media post highlighting talks by Cerebras CTO on GPU programming and AI hardware architecture, noting their low view counts despite being informative.
Cerebras has announced the availability of the Qwen 3.8 27B model on its inference platform with a speed of 1500 tokens per second, detailing its model compression techniques such as quantization and pruning.
The social media post speculates that extremely fast AI models from OpenAI and Cerebras will transform work by enabling parallel agents and text-based computer interactions.
Cerebras announced the CS-5 chip at Hot Chips 2026, offering substantial performance improvements over previous generations with up to 10,000 tokens/sec/user for AI models like Gemma 4 and GPT variants. An AMA session is planned to discuss the announcements.
Cerebras has released the CS-4, a new AI computer that is multiple times faster than its predecessor, claiming a speed advantage over Nvidia systems, with wider availability in the third quarter.
The article explores the 'deadline dividend' concept in AI, where faster inference speeds enable more computation within time constraints, highlighted by OpenAI's Ultrafast API preview for GPT-5.6 Sol and comparisons with providers like Cerebras.
OpenAI paid about $100 to acquire a 4.2% stake in Cerebras through vested warrants, weeks before previewing Ultrafast, a service tier using Cerebras hardware for up to 14 times faster inference of GPT-5.6 Sol.