Award Banner
Award Banner

Trellis Data breakthrough to reduce cost and carbon footprint of AI

Trellis Data breakthrough to reduce cost and carbon footprint of AI Australian AI company develops world-first method to improve speed and reduce computation power requirements of LLMs GlobeNewswire November 11, 2024

CANBERRA, Australia, Nov. 11, 2024 (GLOBE NEWSWIRE) -- Leading machine learning and AI specialist Trellis Data has today announced a groundbreaking innovation in large language model (LLM) decoders that enables outputs to be generated more than three times faster.

Trellis’s Dynamic Depth Decoding (D3) technique uses speculative decoding to achieve an average speed increase of 44% compared with the previous fastest decoder; it is the fastest decoder available in the world today.

LLMs are the architecture behind AI systems, and D3 enables them to generate text approximately 3.16 times faster than standard decoding without compromising on accuracy.

The World Economic Forum estimates the computational power required to sustain global growth in AI is doubling every 100 days. With AI systems creating more and more value for organisations around the world, methods that create efficiencies in cost and energy requirements and mitigate environmental impact have become highly sought after.

According to Trellis Data research submitted for global peer review, the improvement in speed delivered by D3 translates to an estimated 68.4% reduction in computational power requirements, which in turn lowers the costs and carbon emissions associated with running LLMs.

Trellis Data CEO Michael Gately said, “There are three components critical to the success of any AI system: trust, speed and performance. Speed is a key area because it has a direct correlation with cost and environmental impact.

“D3 enables us to address one of the key bottlenecks of speed – the decoder – offering customers a reduction in the cost of running AI servers and a lower carbon footprint.”

“Trellis Data is at the forefront of the rapidly developing AI field and we’re excited to make D3 available on our platform. D3 will provide customers using our speech management and knowledge management capabilities immediate benefits including an enhanced user experience and greater ability to scale.”

D3 is broadly applicable to generative AI use cases, including chatbots, text summarisation, translation, transcription, and content generation. Additionally, the reduced computational costs and carbon emissions deliver a competitive advantage for customers, particularly as LLM size and usage continues to grow.

Trellis Data uses a broad spectrum of AI technologies, including generative AI, transcription, translation, and computer vision, specifically designed to save time and reduce costs. The development of D3 is another key milestone in Trellis Data’s strategy to address the important concerns relating to the adoption of new AI technologies, as they deliver secure, disconnected AI applications that seamlessly integrate with existing business systems.

Headquartered in Canberra, Trellis Data has a growing team of over 40 staff with annual recurring revenue (ARR) growth set to exceed 100%. The company also has offices in Adelaide, Sydney and Arlington (Virginia, USA), with imminent plans to expand into the Middle East and South East Asia.

Trellis Data provides AI solutions to a diverse range of customers, including government agencies, defence and law enforcement, training and supply chain enterprises, and companies in high-security sectors such as finance and legal.

Michael Gately is speaking at MilCIS 2024 in Canberra on 12 November and is a finalist in the Artificial Intelligence category of the 2024 AmCham Alliance Awards, with winners to be announced at a Gala in Sydney on 14 November.

For more information about Trellis Data and its products and services, visit www.trellisdata.com

About Trellis data
Trellis Data is a leading provider of sovereign AI solutions, specialising in the development of secure, disconnected AI applications for the government and defence sectors. With a focus on creating efficiencies and cost savings, Trellis Data leverages cutting-edge technologies in GEN AI, transcription, translation and computer vision to deliver innovative solutions that integrate seamlessly with existing business systems. Headquartered in Canberra, Australia, Trellis Data is committed to driving growth and excellence in the AI ​​industry, both locally and internationally.


Contact Information:

India Bednall
FTI Consulting
M: +61 415 439 800
E: india.bednall@fticonsulting.com

Tim McLaren
Head of Communications
M: +61 402 339 656
E: tim.mclaren@trellisdata.com.au

Primary Logo

This business broadcast service is brought to you by GlobeNewswire through syndication. We have not reviewed or endorsed the content. For any corrections and clarifications, please send it to GlobeNewswire Contact Us Page. If you still require further assistance, please contact our support team at businessbroadcast@asiaone.com.

homepage

trending

trending
    Over 1,800 arrested in cross-border anti-scam ops with losses amounting to $289m
    'Thank you for your hard work': Scoot gives 4.91 months of bonus, shares on TikTok
    Singaporean businessman arrested at Bangkok airport for alleged $200k tax evasion
    Boy, 4, caught smoking under a Bangkok bridge sent to government-run shelter
    HairFun salons admit to targeting, misleading seniors to buy packages after elderly man forced to pay $1,000 for hair treatment
    Paige Chua to perform piano publicly for first time at Singapore Heart Foundation charity show
    'I don't think the situation has been handled very fairly': Parents frustrated as ECDA probes Little Paddington Preschool's closure
    Kym Ng, Carrie Wong, Tay Ying and more to perform for Singapore Heart Foundation 55th Anniversary Charity Show
    'The Labubu I need!' Pop Mart hints at new food-inspired collection, exciting fans
    'Funniest detour I've ever done': Motorcyclist helps buy petrol for stranded rider, wins praise online
    Extension of Woodlands Checkpoint at Old Woodlands Town Centre to begin in September
    Taiwan's popular noodle chain Xiao Hun Mian opens first Singapore outlet at Raffles City

Singapore

Singapore
    • Off-duty cop helps catch thief on Scoot flight, urges public to 'stay vigilant'
    • East-West Line disruption: SMRT to be fined $3m for September 2024 incident
    • 'I feel incredibly honoured': Drum major who dreamed of role as teen leads Singapore Police Force Band in centennial celebration
    • 12-year-old girl locks herself in room, police negotiators called in
    • A year's jail for maid who bit, choked 90-year-old woman with dementia
    • Singapore Pools launches new Toto fixed prize game debuting with June 12 draw
    • 'Both cases likely result of accidents': NParks addresses concerns over community cat deaths
    • Online claims about Covid-19 autopsy and vaccination laws are false: MOH
    • ICA issues verbal advisory to sole voter in Tampines Changkat SMC polling district for failing to update address
    • Beach Road slashing: Man gets 19 years' jail, caning for attempted murder of wife

Entertainment

Entertainment
    • Him Law explains the 'enjoyment' of playing villainous characters
    • Thai actor Nonkul loses mother to cancer shortly before birthday, cancels fanmeet
    • Hong Kong celebrity couple Benjamin Yuen and Bowie Cheung expecting second baby
    • 'I made a fool of myself': Malaysian woman trying to buy G-Dragon concert tickets accidentally buys ones for Kenny G
    • Jackie Chan learnt English through country music
    • James Blunt insists being 1-hit wonder is 'healthy' for his ego
    • Ed Sheeran at top of guest list for Benny Blanco and Selena Gomez's wedding
    • Britain's Prince Harry explored changing surname to Spencer, says Guardian report
    • Travis Kelce celebrates Taylor Swift regaining control of her music
    • Cillian Murphy to feature in The Bone Temple, sequel for 2002 movie 28 Years Later

Lifestyle

Lifestyle
    • Unable to bear children, she proposed annulment of marriage so he could start a family. He chose love
    • New resort chalet run by co-living brand Coliwoo to open in Pasir Ris
    • Miss World 2025 sees first winner from Thailand - meet the political science student who champions breast cancer awareness
    • Jurassic World, inflatable playgrounds and more: Family-friendly events and activities this June holiday
    • What to do this weekend (June 6 to 8)
    • Sushiro to open its first-ever digital conveyor belt system in Singapore at new Mandai outlet
    • COE prices fall across most categories in first bidding exercise for June 2025
    • We asked frequent concertgoers what makes the ultimate concert experience – here's what they said
    • Kyoto's viral Kichi Kichi Omurice chef is coming to Singapore, here's how you can meet him
    • Is Phnom Penh Southeast Asia's most underrated capital? Here's why it is time to visit

Digicult

Digicult
    • Slim, sleek, but slightly too short-lived: Samsung Galaxy S25 Edge review
    • World's best Dota 2 teams to compete for $1m prize pool in Singapore in November
    • A $500 wake-up call: How the Samsung Galaxy Ring made me realise my stress
    • Monster Hunter Wilds producer explains how game has remained unique and fresh over 20 years
    • Initiative by IMDA, AI Verify Foundation tests AI accuracy, trustworthiness in real-world scenarios
    • Under siege? Helldivers 2's latest city to be invaded by aliens could be spoof of Singapore
    • Honor 400 Series launches in Singapore with first free in-device AI image-to-video tool
    • Home Team humanoid robots to be deployed by mid-2027, $100m to be invested: Josephine Teo
    • Ado concert review: Singer without a face ignites fans while in cage with only silhouette visible
    • EU and US authorities take down malware network

Money

Money
    • Wall Street equity indexes close higher after US-China tariff truce
    • Giant deal: Malaysian company to acquire Cold Storage and Giant supermarket chains in Singapore
    • US House plans quick action on Trump cuts to foreign aid spending
    • 4-room HDBs without million-dollar sales - where to still find value today
    • $1.16m for a 4-room HDB flat in Clementi? Why this integrated development commands premium prices
    • Why these buyers chose older leasehold condos — and have no regrets
    • Can you still own multiple properties in Singapore? Here's what you need to know in 2025
    • Selling your home for the first time? Here's a step-by-step timeline to follow in Singapore
    • Why some central 2-bedroom homeowners in Singapore are stuck
    • How the interest rate cycle works - and what it means for your home loan

Latest

Latest
  • China's Sichuan province to offer extended marriage and maternity leave
  • Pressure mounts on Netanyahu as opposition moves to dissolve parliament 
  • China issues bounty for hackers it says are linked to Taiwan 
  • At least 11 die in crowd chaos outside Indian cricket stadium
  • Hundreds of Venezuelans deported to El Salvador have right to challenge detention, US judge rules 
  • Southwest Pacific hit by unprecedented marine heat waves in 2024, UN says 
  • Russian drones strike dwellings in Ukraine's Kharkiv, injure 17, governor says 
  • CDC official resigns from Covid vaccine committee advisory role, sources say 
  • Trump signs proclamation to restrict student visas at Harvard 

In Case You Missed It

In Case You Missed It
  • Over 170 travellers nabbed for evading GST, smuggling large sums of cash in island-wide operation
  • Woman crawls out of storm drain in shocking Manila street scene
  • DBS staff, police stop 2 victims from losing $820k to government official impersonation scams
  • 'Be humble in victory': PM Wong sends traditional 'Rules of Prudence' letter to PAP MPs after GE
  • Pedestrian, 84, dies in accident involving minibus in Choa Chu Kang
  • NDP 2025 marks SG60 with expanded celebrations from Padang to Marina Bay
  • Obesity rates are rising in Singapore, but is overeating the only cause?
  • Trump administration blocks Harvard from enrolling foreign students, threatens broader crackdown 
  • 'We apologise for the operational lapse': NUS responds to backlash over disposal of Yale-NUS books
This website is best viewed using the latest versions of web browsers.