Intel Corporation (INTC) announced significant advancements in artificial intelligence (AI) inference capabilities, revealing performance gains during the MLPerf Inference v6.1 benchmarks. The Intel Xeon 6980P processors achieved 2.4 times higher throughput for Llama 3.1 (8B Server) and a 56% increase in Offline throughput compared to MLPerf v6.0, reflecting effective software optimization. Intel expanded its Xeon 6 participation from two to five processor models, boosting CPU inference results from 24 to 35, along with an increase in partners participating in the benchmarks from 29 to 39.
Additionally, Intel’s Arc Pro B70 GPUs supported major workloads, including Llama 2 70B and gpt-oss-120B, with a four-GPU system showing 128GB of VRAM and performance improvements of 36% in Server and 27% in Offline modes compared to previous results. The company also co-developed results for the end-to-end retrieval-augmented generation benchmark, showcasing a combined system of a Xeon 6787P processor and four Arc Pro B70 GPUs.
Shares of Intel rose 230.6% over the past year, significantly outperforming the industry average growth of 27.1%. The company currently has a Zacks Rank #3 (Hold), with earnings estimates for 2026 increasing by 44.2% to $1.50 per share and by 30.7% to $1.96 for 2027 over the past two months.
5 Stocks Our Experts Predict Could Double In the Next Year
By submitting your email, you'll also get a free pivot & flow membership. A free daily market overview. You can unsubscribe at any time.
