Amazon offers $110 million in credits for university AI research

$110 million in university AI research support

Amazon has unveiled the 34 recipients of the Fall 2025 cycle of its "Build on Trainium" program, a $110 million computing credit initiative designed to support AI research and university education using AWS Trainium infrastructure. This round's recipients are teams that submitted research proposals in five priority areas under the Responsible AI theme: AI safety and alignment, multilingual language models, representation engineering, sustainability and small language models, and synthetic data generation.

Amazon said proposals were evaluated based on the quality of their scientific content and their potential impact on the research community and society.

Resources provided to recipient researchers

Through AWS promotional credits, recipient researchers gain access to more than 700 public Amazon datasets and can use AWS AI/ML services and tools. Each team is also assigned an Amazon research staff member for guidance and advice, and tutorials and hands-on sessions related to AWS Trainium are provided.

Yida Wang, Principal Applied Scientist at AWS AI, explained that the program gives researchers robust, scalable access to Amazon's custom AI chips. He noted that a UIUC team is studying topology-aware parallelization strategies for mixture-of-experts models with up to a trillion parameters across as many as 1,024 Trainium chips, while a University of Washington team is developing an inference optimization framework to improve LLM inference performance on Trainium.

List of research selected in this cycle

The 34 selected research teams this round come from a range of universities including UIUC, UCLA, CMU, MIT, Johns Hopkins, UC San Diego, Northwestern, and Georgia Tech. Research topics span a wide range of areas, including federated adversarial co-training for LLM security and robustness, hybrid public-private approaches to differentially private synthetic data generation, multimodal scam content detection and mitigation, algorithm-system co-design for sparse and quantized LLMs, multilingual speech-to-text large language models, safe pretraining for LLM agents, memorization-aware preference optimization for machine unlearning, multi-stage and multi-objective alignment for multilingual language models, synthetic data scaling for pretraining low-resource language models, certified robustness training for quantization- and pruning-aware vision-language models, LLM hallucination detection and mitigation, and safety benchmarking for tool use by LLM agents.

The full list of recipients can be found on Amazon Science's official website.

How to join the program and its constraints

Build on Trainium operates through an open call for proposals aimed at university research teams. This round's recipients were selected from the Fall 2025 cycle's call for proposals. The source did not disclose specific details about the schedule or application process for the next cycle. Program details should be checked separately on the official Amazon Research Awards page.