#semiconductors

Intel bet everything on x86 dominance, and now its foundry arm is flailing while AI chips move to entirely new architectures. The lesson isn't about manufacturing-it's about the perils of betting your future on a single, rigid design philosophy. Diversify or die applies to chip design, too. #intel #semiconductors #chipdesign

Olix, a photonics startup, has raised $220 million in Series B funding to develop commercial AI chips that process data using light rather than electricity. The company claims its architecture reduces latency thousands of times over, cuts data center energy consumption by 80-90%, and enables parallel computing beyond silicon's physical limits. If successful by late 2026, the cost of training large language models could drop dramatically. The investment signals growing interest in alternatives to Nvidia's semiconductor dominance. https://vir.group/ #TechNews #AI #Semiconductors

Toronto-based Taalas just emerged from stealth with a claim that’s shaking the hardware world: 17,000 tokens per second on Llama 3.1 8B. How? By physically etching the AI model directly into the silicon transistors. No HBM. No liquid cooling. Just raw, hardwired performance that is 10x faster and 20x cheaper than traditional GPU inference. https://www.buysellram.com/blog/17000-tokens-second-is-taalas-hardwired-silicon-the-ultimate-solution-to-the-ai-memory-wall-and-hbm-shortage/ The Breakthrough: Taalas has unveiled the HC1 chip, achieving a massive 17,000 tokens/second on Llama 3.1 8B. It is roughly 10x faster and 20x cheaper than traditional GPU inference. The “Hardwired” Secret: Unlike GPUs that load software, Taalas etches the AI model directly into the silicon transistors. By physically embedding the weights, they eliminate the need for High-Bandwidth Memory (HBM). Solving the Memory Wall: By removing the “data movement” between external memory and the processor, Taalas bypasses the industry’s biggest bottleneck—the Memory Wall—and operates entirely on standard air cooling. The Trade-off: The chip is model-specific. While it offers “insane” efficiency for stable, high-volume production (like 24/7 chatbots), it lacks the programmability and flexibility of a GPU. Market Impact: The rise of these specialized “Inference Factories” actually increases the long-term value of your GPUs. Because GPUs are versatile and can be repurposed for any new model, they remain the “Gold Standard” for resale and training. Demo LLM: chat jimmy #AI #ArtificialIntelligence #Hardware #Semiconductors #DataCenter #MemoryWall #HBMShortage #InferenceFactory #HardcoreAI #ASIC #Taalas #Llama3 #NVIDIA #technology

HCLTech and Suchi Semicon have partnered to offer an integrated design-to-backend semiconductor solution: HCLTech brings design expertise and global customer access, while Suchi provides OSAT manufacturing in Gujarat. Unlike HCLTech's Foxconn venture, this partnership involves no equity participation. https://economictimes.indiatimes.com/tech/technology/hcltech-suchi-semicon-team-up-to-offer-semicon-solutions/articleshow/128825402.cms #India #Tech #News #Semiconductors #HCLTech

Four of 10 semiconductor fabrication and packaging projects approved under India's Semiconductor Mission will begin production very soon, PM Modi has announced. Speaking at the groundbreaking of HCL-Foxconn's OSAT facility in Jewar, Modi said India strives to join top nations in providing processing power to the world. https://economictimes.indiatimes.com/tech/technology/four-semicon-units-to-begin-production-very-soon-says-pm-modi/articleshow/128678577.cms #India #Tech #News #Semiconductors #Manufacturing