OpenAI’s 700W Jalapeño ASIC outpaces 1,400W Nvidia flagship GPU — claims up to 1.9x throughput per kilowatt and 3.6x lower latency, co-developed with Broadcom
Source: Tom's Hardware UK· Luke James· August 25, 2026
SynaBot summary
OpenAI revealed its custom 700W "Jalapeño" AI chip, developed with Broadcom, boasting superior performance per watt and lower latency compared to Nvidia's top-tier GPUs. This in-house silicon aims to enhance OpenAI's training and inference capabilities.
Key takeaways
- OpenAI's custom chip offers better energy efficiency.
- Lower latency promises quicker AI task completion.
- In-house hardware could reduce reliance on external vendors.
- This challenges Nvidia's established AI hardware market share.
Why it matters
This development signals a potential shift in AI hardware dominance, impacting the cost and efficiency of running AI models. For users, it could mean faster AI responses and more accessible, powerful AI tools as specialized hardware becomes more competitive.
This story was reported by Tom's Hardware UK. Read the full original article:
Read on Tom's Hardware UK Try this on SynaBot
Related AI assistants, prompts, and tools from the SynaBot catalog.
AI Tools
- OpenAI CodexOpenAI Codex is a large language model fine-tuned for programming, capable of translating natural language into code across multiple programming languages. It powers tools like GitHub Copilot.
- OpenAI SoraOpenAI's Sora is a text-to-video model capable of generating long, complex scenes with multiple characters, specific types of motion, and accurate subject and background details. It represents a significant leap in generative video AI.
- ChatGPT by OpenAIDeveloped by OpenAI, ChatGPT is a highly capable conversational AI that generates human-like text based on prompts. It can answer questions, write essays, summarize documents, and engage in creative dialogue across a vast range of topics.
