Research lab
Allen Institute for AI, a Seattle nonprofit research lab that builds fully open language models (Olmo), training tools, benchmarks and science agents.
Ai2 says its HiRO-ACEv2 system generates global 3 km climate data at about 1/300th the cost of the X-SHiELD model. The 3% match on extreme-event statistics came from a narrower test, with the downscaler fed coarsened X-SHiELD data.
Reality
Ai2's Olmo-core 3 reached about 52,000 tokens per second per GPU on a 47B mixture-of-experts model, roughly 2.7 times the old stack, on eight B300 GPUs. The report labels every headline rate as measured under random routing, so a team with a trained router has to rerun the comparison.
Reality
Robocurve's RoboHarm harness put three frontier policies through 300 fixed trials on a pair of $2,999 arms. The published per-trial logs show where a refusal landed and what it cost in model calls.
Perspective Coverage
4 publishersReality
Ai2 replaced priority scheduling on thousands of GPUs with time budgets and hierarchical fair-share after every scheduled workload ended up marked HIGH. The change moves the argument over who gets GPU time out of case-by-case operations and into an administrative budget.
Reality
Ai2 released Olmo-core 3, an open mixture-of-experts training stack it has benchmarked at over one trillion total parameters. Its speedups are measured against Ai2's own earlier FSDP code, so teams on Megatron-Core must run their own comparison.
Perspective Coverage
3 publishersReality
Ai2 open-sourced AstaBrief 8B, which writes cited research reports in 51.1 seconds against 178.5 for Asta's Claude-powered mode. Labs that cannot send unpublished research questions to a hosted model can now run a cited-report generator on their own servers.
Perspective Coverage
3 publishersReality
Nathan Lambert and Tom Zick launched Trillium Labs, a nonprofit that will publish AI experiments, self-improvement work included, for outsiders to replicate. Their bet is that outside scrutiny will find and limit frontier-model risks better than keeping models locked inside labs.
Reality
Ai2 fit a multidimensional item response model to 100 models across 16 benchmarks. The per-question estimates say parts of its own safety suite are grading general reasoning rather than refusal behaviour.
Reality