Cerebras
AssessPlatforms
An AI compute platform and hardware company focused on high-speed model inference and training.
Why it's here
Placed in Assess: 8 article(s) of evidence from 3 source(s), led by product launches, with 0 in the last 30 days. Confidence 69%.
Evidence (8)
- 4Hacker News·9/3/2026model_releaseCerebras adds Qwen 3.8 27B to public endpoints
Cerebras says Qwen 3.8 27B is now available on its public endpoints at about 1,500 tokens per second, alongside OpenAI GPT OSS 120B. The page also clarifies that public models are served unpruned, with selective weight-only quantization used for storage and full-precision activations, attention, and KV cache.
- 10Hacker News·9/1/2026securityFBI investigates dark web service selling 153M+ driver’s licenses
A dark web service called Nexus is reportedly selling digital scans of more than 153 million driver’s licenses and other identity documents from the U.S. and Canada. KrebsOnSecurity says the FBI’s New Orleans field office has opened an inquiry into the source of the images, which may come from a breach at an identity verification company.
- 7Hacker News·8/23/2026researchSurvey of AI Chip Architectures
This article reviews the rise of domain-specific AI chip architectures in response to the memory wall and the limits of Moore’s Law. It compares deployment approaches such as GPUs, TPU-style systolic accelerators, wafer-scale engines, and low-latency inference chips, focusing on how they move data and scale for training and inference.
- 8Hacker News·8/19/2026product_launchCerebras launches CS-4 rack-scale AI inference system
Cerebras announced the CS-4, a new rack-scale AI inference platform built on its WSE-3 Turbo wafers and Nexus Rack-Scale Platform. The company says it can deliver up to 30x faster inference than GPU systems, with higher throughput per watt and deployment times reduced from days to hours.
- 8Hacker News·8/13/2026product_launchOpenAI and Cerebras Launch Ultrafast Mode for GPT-5.6 Sol
Cerebras and OpenAI announced Ultrafast Mode, a new OpenAI API tier powered by Cerebras for GPT-5.6 Sol. The service is designed to deliver very high output speed, claiming up to 750 tokens per second without quality loss, and is initially available to a limited set of customers. The companies say the mode improves latency for time-sensitive workloads such as incident response, security operations, and agentic workflows, and cite benchmark results showing substantially faster execution on knowledge-work tasks.
- 7Hacker News·7/8/2026model_releaseCognition Launches SWE-1.7 Coding Model
Cognition announced SWE-1.7, its latest coding model, claiming it reaches frontier-level performance at lower cost and improves on prior results in agentic software engineering tasks. The company says the model was trained with RL pipeline improvements, better data quality, multi-cluster training, and self-compaction for long-horizon tasks, and is now available in Devin through Cerebras.
- 7Hugging Face Blog·7/1/2026product_launchHugging Face and Cerebras enable real-time Gemma 4 voice AI
Hugging Face and Cerebras showcased a real-time speech-to-speech demo built around Gemma 4 31B, aiming to reduce latency in voice AI. The open modular pipeline combines Parakeet for speech recognition, Cerebras for fast model inference, and Qwen3TTS for text-to-speech, and is already used in Reachy Mini robots.
- 8OpenAI Blog·1/14/2026product_launchOpenAI adds Cerebras compute for faster ChatGPT inference
OpenAI is partnering with Cerebras to add 750MW of high-speed AI compute capacity. The move is intended to reduce inference latency and improve ChatGPT performance for real-time workloads.