Skip to main content

OpenAI launches Jalapeño, its first custom inference chip

Source Decision Insights Record Published October 25, 2026 Companies 11
Human View Machine View

11 companies named across 3 categories, one of 75 articles referencing OpenAI. Previous coverage: NVIDIA introduces OpenShell reference architecture for agents (Oct 2026).

Companies mentioned

Best suited for

Seniority
C Level / Executive Team
Job function
Chief Technology Officer
Persona
Platform / Infrastructure Architect
Buyer role
Architect / Technical Evaluator
Decision stage
Where to buy
Adoption curve
Early Adopters
Technology maturity
Accelerated Attention
Industry
Information Technology / AI Infrastructure / AI Compute & Systems / AI Accelerators (GPUs & AI ASICs)

Our classification, not the publisher's statement. Best suited for, not only for.

OpenAI has moved its first custom AI chip, Jalapeño, from a development partnership to working silicon. The chip extends the company's work beyond models and software into hardware, with the aim of lowering the cost of inference across its services. The step adds to a wider pattern of AI developers and cloud operators designing their own processors.

The effort builds on an October 2025 agreement with Broadcom that set up a multiyear custom-chip program worth $10 billion, covering 10 gigawatts of OpenAI-designed accelerators and networking systems. Google, Amazon, Microsoft and Meta have already built proprietary accelerator programs, and Anthropic has started assembling an in-house chip-design team.

Jalapeño is designed around large language model inference rather than adapting a general-purpose GPU architecture. Its design focuses on minimizing data movement and balancing compute, memory and networking resources. OpenAI designed the architecture, Broadcom supplies silicon implementation and networking technology, Celestica supports board and rack integration, and TSMC will manufacture the chips.

Engineering samples are running machine-learning workloads in the laboratory at production-target frequency and power, and OpenAI plans an initial deployment in small volumes near the end of 2026. In August 2026, the company published initial benchmarks claiming performance-per-watt and latency advantages over competing systems. These figures are company-reported and do not yet establish a sustained production-scale advantage over Nvidia's latest accelerators.

Anthropic began forming its chip-design team in August 2026 to develop hardware optimized for Claude, but it has not announced a production-ready proprietary accelerator or deployment timetable. Designing advanced chips requires substantial capital, specialized engineering talent and access to manufacturing and packaging capacity. The central question is whether Jalapeño can deliver lower total cost per AI token at production scale. Whether that outcome materializes remains to be seen.

Press release, originally published by Alan Weissberger at techblog.comsoc.org.

Mentions

Structured data (JSON-LD)

The schema.org markup this page publishes for search engines and AI agents, exactly as they read it.

[
  {
    "@context": "https://schema.org",
    "@id": "https://decisioninsights.ai/#website",
    "@type": "WebSite",
    "name": "Decision Insights",
    "potentialAction": {
      "@type": "SearchAction",
      "target": {
        "@type": "EntryPoint",
        "urlTemplate": "https://decisioninsights.ai/search/?q={search_term_string}&submit=1"
      }
    },
    "publisher": {
      "@id": "https://decisioninsights.ai/#organization"
    },
    "url": "https://decisioninsights.ai"
  },
  {
    "@context": "https://schema.org",
    "@id": "https://decisioninsights.ai/#organization",
    "@type": "Organization",
    "contactPoint": {
      "@type": "ContactPoint",
      "contactType": "customer support",
      "email": "[email protected]"
    },
    "description": "Decision Insights is a Registry of technology companies, open source projects, and industry terms, built for people and for AI agents that need sourced, structured information.",
    "legalName": "Wiretap Labs, LLC",
    "logo": {
      "@type": "ImageObject",
      "url": "https://wiretap-cdn-assets.nyc3.cdn.digitaloceanspaces.com/decision-insights/[email protected]"
    },
    "name": "Decision Insights",
    "parentOrganization": {
      "@type": "Organization",
      "name": "Wiretap Labs",
      "sameAs": [
        "https://www.linkedin.com/company/wiretap-labs",
        "https://www.crunchbase.com/organization/wiretap-labs"
      ],
      "url": "https://wiretaplabs.com"
    },
    "publishingPrinciples": "https://decisioninsights.ai/standards/",
    "sameAs": [
      "https://www.linkedin.com/company/decisioninsights"
    ],
    "url": "https://decisioninsights.ai"
  },
  {
    "@context": "https://schema.org",
    "@type": "BreadcrumbList",
    "itemListElement": [
      {
        "@type": "ListItem",
        "item": "https://decisioninsights.ai",
        "name": "Decision Insights",
        "position": 1
      },
      {
        "@type": "ListItem",
        "item": "https://decisioninsights.ai/records/",
        "name": "Records",
        "position": 2
      },
      {
        "@type": "ListItem",
        "item": "https://decisioninsights.ai/openai-launches-jalapeo-its-first-custom-inference-chip/",
        "name": "OpenAI launches Jalapeño, its first custom inference chip",
        "position": 3
      }
    ]
  },
  {
    "@context": "https://schema.org",
    "@id": "https://decisioninsights.ai/openai-launches-jalapeo-its-first-custom-inference-chip/#blogposting",
    "@type": "NewsArticle",
    "about": {
      "@id": "https://decisioninsights.ai/registry/openai/#organization",
      "@type": "Organization",
      "mainEntityOfPage": "https://decisioninsights.ai/registry/openai/",
      "name": "OpenAI"
    },
    "articleSection": [
      "Networking",
      "Silicon"
    ],
    "audience": [
      {
        "@type": "Audience",
        "additionalType": "Seniority",
        "audienceType": "C Level / Executive Team"
      },
      {
        "@type": "Audience",
        "additionalType": "Job function",
        "audienceType": "Chief Technology Officer"
      },
      {
        "@type": "Audience",
        "additionalType": "Persona",
        "audienceType": "Platform / Infrastructure Architect"
      },
      {
        "@type": "Audience",
        "additionalType": "Buyer role",
        "audienceType": "Architect / Technical Evaluator"
      },
      {
        "@type": "Audience",
        "additionalType": "Adoption curve",
        "audienceType": "Early Adopters"
      },
      {
        "@type": "Audience",
        "additionalType": "Technology maturity",
        "audienceType": "Accelerated Attention"
      },
      {
        "@type": "Audience",
        "additionalType": "Industry",
        "audienceType": "Information Technology / AI Infrastructure / AI Compute & Systems / AI Accelerators (GPUs & AI ASICs)"
      }
    ],
    "author": {
      "@id": "https://decisioninsights.ai/author/decision-insights-signals/#person",
      "@type": "Person",
      "name": "Decision Insights Record",
      "url": "https://decisioninsights.ai/author/decision-insights-signals/"
    },
    "dateModified": "2026-10-25T19:32:42-06:00",
    "datePublished": "2026-10-25T19:32:36-06:00",
    "description": "OpenAI's Jalapeño custom AI inference chip has reached working silicon, with initial deployment planned for late 2026 and expansion over multiple generations.",
    "headline": "OpenAI launches Jalapeño, its first custom inference chip",
    "isBasedOn": {
      "@type": "CreativeWork",
      "author": {
        "@type": "Person",
        "name": "Alan Weissberger"
      },
      "sourceOrganization": {
        "@id": "https://decisioninsights.ai/registry/ieee/#organization",
        "@type": "Organization",
        "mainEntityOfPage": "https://decisioninsights.ai/registry/ieee/",
        "name": "IEEE"
      },
      "url": "https://techblog.comsoc.org/2026/10/25/custom-ai-silicon-openai-anthropic-hyperscalers-and-the-battle-with-nvidia-amd-and-broadcom/"
    },
    "keywords": [
      "Inference"
    ],
    "mainEntityOfPage": {
      "@id": "https://decisioninsights.ai/openai-launches-jalapeo-its-first-custom-inference-chip/",
      "@type": "WebPage",
      "sdDatePublished": "2026-10-26",
      "sdPublisher": {
        "@id": "https://decisioninsights.ai/#organization"
      }
    },
    "mentions": [
      {
        "@id": "https://decisioninsights.ai/registry/amazon-com/#organization",
        "@type": "Organization",
        "mainEntityOfPage": "https://decisioninsights.ai/registry/amazon-com/",
        "name": "Amazon"
      },
      {
        "@id": "https://decisioninsights.ai/registry/anthropic/#organization",
        "@type": "Organization",
        "mainEntityOfPage": "https://decisioninsights.ai/registry/anthropic/",
        "name": "Anthropic"
      },
      {
        "@id": "https://decisioninsights.ai/registry/broadcom/#organization",
        "@type": "Organization",
        "mainEntityOfPage": "https://decisioninsights.ai/registry/broadcom/",
        "name": "Broadcom"
      },
      {
        "@id": "https://decisioninsights.ai/registry/celestica/#organization",
        "@type": "Organization",
        "mainEntityOfPage": "https://decisioninsights.ai/registry/celestica/",
        "name": "Celestica"
      },
      {
        "@id": "https://decisioninsights.ai/registry/facebook/#organization",
        "@type": "Organization",
        "mainEntityOfPage": "https://decisioninsights.ai/registry/facebook/",
        "name": "Meta Platforms"
      },
      {
        "@id": "https://decisioninsights.ai/registry/microsoft/#organization",
        "@type": "Organization",
        "mainEntityOfPage": "https://decisioninsights.ai/registry/microsoft/",
        "name": "Microsoft"
      },
      {
        "@id": "https://decisioninsights.ai/registry/nvidia-corp/#organization",
        "@type": "Organization",
        "mainEntityOfPage": "https://decisioninsights.ai/registry/nvidia-corp/",
        "name": "Nvidia"
      },
      {
        "@id": "https://decisioninsights.ai/registry/openai/#organization",
        "@type": "Organization",
        "mainEntityOfPage": "https://decisioninsights.ai/registry/openai/",
        "name": "OpenAI"
      },
      {
        "@id": "https://decisioninsights.ai/registry/services/#organization",
        "@type": "Organization",
        "mainEntityOfPage": "https://decisioninsights.ai/registry/services/",
        "name": "services"
      },
      {
        "@id": "https://decisioninsights.ai/registry/saudi-telecom/specialized/#organization",
        "@type": "Organization",
        "mainEntityOfPage": "https://decisioninsights.ai/registry/saudi-telecom/specialized/",
        "name": "specialized"
      },
      {
        "@id": "https://decisioninsights.ai/registry/tsmc/#organization",
        "@type": "Organization",
        "mainEntityOfPage": "https://decisioninsights.ai/registry/tsmc/",
        "name": "Taiwan Semiconductor Manufacturing Company"
      }
    ],
    "publisher": {
      "@id": "https://decisioninsights.ai/#organization"
    }
  },
  {
    "@context": "https://schema.org",
    "@id": "https://decisioninsights.ai/author/decision-insights-signals/#person",
    "@type": "Person",
    "description": "Press releases and independent analyst research, structured into concise summaries for research and monitoring. Produced under our Standards & Methodology.",
    "name": "Decision Insights Record",
    "sameAs": [
      "https://www.linkedin.com/showcase/decisioninsights/"
    ],
    "url": "https://decisioninsights.ai/author/decision-insights-signals/"
  }
]

Is this your company? Get structured data for your own pages