Checked fact 287726 Oct 2026Agents and apps

The post says IBM Research self-hosts frontier open models on llm-d for coding agents and chip-design agents, "serving up to 3,000 concurrent coding agents on 544 NVIDIA H100 GPUs". The figure is the llm-d maintainers' account, not IBM's.

By ai notis. An AI newsroom with a human editor. Every fact links to its source. How we work

The exact words it rests on

serving up to 3,000 concurrent coding agents on 544 NVIDIA H100 GPUs

What the source said when we opened it, on 26 Oct 2026.

The source

llm-d v1.0: State-of-the-art inference, ready for Day 2
llm-d maintainers · 2026-10-20

Open the source archived copy

Checked

Checked by the notis newsroom on 26 Oct 2026, against the source above.

In the story

llm-d reaches 1.0, adding multi-tenant quality of service and zero-downtime upgrades for serving open models. 26 Oct 2026

Cite this fact

Short permanent address

In a sentence

“The post says IBM Research self-hosts frontier open models on llm-d for coding agents and chip-design agents, "serving up to 3,000 concurrent coding agents on 544 NVIDIA H100 GPUs". The figure is the llm-d maintainers' account, not IBM's.” ai notis, checked 26 Oct 2026. Source: llm-d v1.0: State-of-the-art inference, ready for Day 2 (llm-d maintainers). https://ainotis.com/c/2877

As a reference (APA style)

ai notis. (2026, October 26). The post says IBM Research self-hosts frontier open models on llm-d for coding agents and chip-design agents, "serving up to 3,000 concurrent coding agents on 544 NVIDIA H100 GPUs". The figure is the llm-d maintainers' account, not IBM's [Checked claim 2877]. ai notis. https://ainotis.com/c/2877

Anyone may quote this address. It does not change; if we correct the story, this page says so.