AMD experimental figures report automated porting of 51 models to Strix Halo

AMD has released experimental figures showing an automated AI process ported 51 models to its Strix Halo platform in four days. The company reports that reading long text is faster on the hardware, with processing 32,000 tokens reaching a median speedup of 1.5 times. Maximum processing performance reportedly reached approximately 2.1 times faster in test conditions.

Faster processing for large token windows could improve local long-context inference for developers targeting edge silicon. However, these metrics come solely from early vendor benchmarks and require independent validation under realistic production workloads.

  • An AI workflow ported 51 models to Strix Halo within four days, according to AMD.
  • AMD experimental figures show long-context processing of 32,000 tokens is 1.5 times faster at the median.
  • Reported performance gains for reading long text reached up to approximately 2.1 times faster in experimental tests.
Machine Learning Generative AI Sovereign AI
All AI news

More AI news

Models

OpenAI releases GPT-6.2 Astra with lower API pricing

OpenAI has introduced GPT-6.2 Astra alongside a 20 percent reduction in API pricing. According to the organisation, the release marks its first publicly available model to attain a High rating on its internal evaluation framework.

Models

OpenAI model GPT-6.2 Astra Max enters SevenLab leaderboard at third place

OpenAI's GPT-6.2 Astra (Max) has entered the top 10 of the SevenLab AI leaderboard at rank number three. The position is calculated using SevenLab's value-adjusted ranking methodology, which draws upon benchmark data from ArtificialAnalysis.

Industry

WhiteMatter raises 20 million dollars in series A funding for AI guardian agents

WhiteMatter announced a 20 million dollar Series A funding round on 10 November 2026. The investment will support the development of software designed to supervise other AI agents.

Build it with us, or have us check yours

In 15 minutes we talk through what you want to build, who builds it and what it needs to go live.

Talk directly with our AI specialists

15 minutes, no obligation
No sales pressure
Prototype in 7 days