Broadcom and OpenAI Unveil Jalapeño LLM Inference Chip
TL;DR
- Broadcom CEO Hock Tan publicly cited roughly 50% cost savings vs. typical AI GPUs, the first concrete cost figure from either company.
- OpenAI's own AI models accelerated chip design and optimization, making Jalapeño a recursive product: AI designed to run AI faster.
- Jalapeño is an ASIC tuned narrowly for LLM inference, trading adaptability for cost and efficiency at scale.
Building a chip from scratch specifically for large-language-model inference, rather than adapting an existing general-purpose accelerator, is a deliberate architectural bet. According to reporting on StockTitan, OpenAI and Broadcom have done exactly that with Jalapeño, which OpenAI describes as "OpenAI's first Intelligence Processor: an accelerator architected around OpenAI's vision for the future of LLM inference."
The chip's design philosophy centers on reducing data movement and balancing compute, memory, and networking resources to achieve utilization much closer to theoretical peak performance. The companies describe it as a blank-slate design for modern LLM inference, not a general-purpose accelerator adapted from earlier AI workloads. Engineering samples are reportedly already running ML workloads in the lab at production target frequency and power, including GPT-5.3-Codex-Spark.
The development pace is the figure that stands out most. The companies say Jalapeño went from initial design to manufacturing tape-out in nine months, which they describe as potentially the fastest ASIC development cycle ever achieved in high-performance advanced semiconductors. That timeline compression, if it proves repeatable, changes what it means to iterate on custom silicon at hyperscaler speed.
Deployment is targeted at gigawatt-scale data centers with Microsoft and other partners beginning in 2026. The honest caveat is that the performance-per-watt claims, described as substantially better than current state-of-the-art, are the companies' own, with no independent benchmarks published yet. What the reporting also does not give you is a precise unit count, pricing structure, or a quantitative comparison against competing silicon on real production workloads.
Broadcom, as the co-designer and manufacturing partner, stands to benefit as custom ASIC demand from frontier model operators grows. Shares of AVGO dipped roughly 3% on the announcement day, though reporting attributed the move to broader semiconductor sector weakness rather than to reception of the chip itself.
What others are reporting
-
OpenAI Read →
First-party post covering the multi-generation platform scope, the AI-assisted design process, and OpenAI's framing of Jalapeño as the first chip in a long-term silicon partnership with Broadcom.
-
Bloomberg Read →
The only Tier-1 outlet to surface Hock Tan's cost savings claim, grounding the announcement in financial terms rather than performance framing alone.
The accelerator is showing cost savings of roughly 50% compared with typical AI graphics processing units.
-
TechCrunch Read →
Focuses on full-stack vertical integration across chip architecture, kernels, memory movement, and scheduling, with Greg Brockman quoted on the workload-specificity rationale.
We have a deep understanding of the workload. We've really been looking for specific workloads that are underserved.
-
Yahoo Finance Read →
Frames Jalapeño as a direct Nvidia challenge and situates it within the broader industry shift, with Amazon, Google, Microsoft, and Meta all building custom silicon to reduce GPU dependency.
By designing more of the stack ourselves, we can serve more intelligence with greater efficiency.
-
The Decoder Read →
The only source to flag that performance claims lack independent verification, with benchmarking conditions unclear and a full technical report still pending.
OpenAI says the architecture delivers better performance per watt. Development took just nine months.
Originally reported by stocktitan.net
Read the original article →Original headline: OpenAI and Broadcom Unveil Jalapeño — First LLM-Optimized Inference Chip Built From Blank Slate in Nine Months, Engineering Samples Running GPT-5.3-Codex-Spark Production Workloads