Skip to content
Lanseringar· LaunchAvailable

OpenAI Unveils Jalapeño AI Chip: Faster and More Energy-Efficient

OpenAI has showcased the first benchmark results for its proprietary AI chip, Jalapeño. The chip delivers up to 1.9 times greater energy efficiency and 3.6 times lower latency for inference tasks.

By the Aheadline editorial team·28 aug. 2026·2 min read·Source: Entity-watch: OpenAIVerifierad signalAI-generated
OpenAI Unveils Jalapeño AI Chip: Faster and More Energy-Efficient
OpenAI Unveils Jalapeño AI Chip: Faster and More Energy-Efficient
OpenAI Unveils Jalapeño AI Chip: Faster and More Energy-Efficient
By · Verktygs- & infrastrukturreporter
Last updated

What happened?

OpenAI has published the initial test results for its proprietary AI inference chip, Jalapeño. The chip demonstrates an increase of up to 1.9 times more work performed per watt and up to 3.6 times lower latency compared to existing hardware solutions. The company developed the chip in collaboration with Broadcom and presented the findings at the Hot Chips architecture conference.

Key facts

Chippets namnJalapeño
Effektivitetsökning per wattUpp till 1,9x
LatensminskningUpp till 3,6x lägre
Planerad utrullningSlutet av 2026

Why it matters

Inference—the phase where pre-trained AI models generate responses for users—accounts for the majority of data centre energy consumption and operational costs. By optimising its own hardware, OpenAI can significantly reduce costs while improving the performance of its large-scale AI services.

Who is affected?

The development primarily affects developers and companies utilising OpenAI’s infrastructure and APIs, as lower latency results in faster response times for AI applications. In the long term, the broader market for AI hardware and cloud services will also be affected as OpenAI reduces its reliance on external chip suppliers.

Impact on the EU

While the proprietary chip does not directly impact the availability or regulation of AI models within the EU, lower operational costs and enhanced energy efficiency may assist OpenAI in meeting EU requirements regarding sustainability and data centre energy consumption.

What else you should know

Tests have been conducted on several notable models, including GPT-OSS 120B, DeepSeek R1, and Kimi K2.5 1T. With these benchmark results, measured in part via SemiAnalysis InferenceX, OpenAI is making a direct entry into the hardware market to challenge established players.

Frequently asked questions

Quick answers about this story

Vad har hänt?
OpenAI har visat upp testresultat för sitt första egna AI-inferenschip Jalapeño, som uppvisar stora förbättringar i hastighet, latens och energieffektivitet.
När hände det?
OpenAI presenterade benchmark-resultaten för Jalapeño tisdagen den 25 augusti 2026 under branschkonferensen Hot Chips.
Varför spelar det roll?
Genom egen hårdvara kan OpenAI kapa driftkostnader för storskalig AI-inferens, erbjuda snabbare svarstider till användare och minska sitt beroende av externa chiptillverkare.
När tas Jalapeño i bruk?
Chippet planeras att tas i bruk i OpenAI:s egen beräkningsinfrastruktur före utgången av 2026.
Original source
Entity-watch: OpenAI·freepressjournal.in

The link opens in a new window and leads to the publisher's own site.

Verifierad signal

Källan har spårats automatiskt från utgivaren via Aheadlines signalkedja.

AI-verktyg i artikeln

Topics

#GPU#AI-benchmarking#AI-chip#AI-inferens
[ STAY UP TO DATE ]

Get similar news straight to your inbox

No affiliate linksCancel anytimeGDPR-friendly
[ Frequency ]
[ What do you want to read about? ]

You'll receive updates on 2 topics.

The reader's room

Send in a question or an addition. The newsroom reads everything before it's published and replies when relevant. No AI-generated text – just people.

Sign in to submit a comment or question.

Loading comments…
How this affects you

Read the article through your role

  • Decide whether this affects strategy over 6–12 months or is just noise.
  • Discuss with leadership: do we own the right question or does ownership need to move?
  • Ask: what risk are we taking by NOT acting on this this quarter?

Generated angle — not editorial analysis of "OpenAI Unveils Jalapeño AI Chip: Faster and More Energy-Effi"