Blog

PrimaLabs Blog Posts

Company news, partnerships, and inference engineering notes from the team.

22 Oct 2026PrimaLabs Engineering

NVIDIA Nemotron. Lightning-fast on PrimaLabs.

Over 500 tokens per second. PrimaLabs ranks #1 for output speed on both Nemotron Super VL and Lightning in Artificial Analysis’s 10k-input benchmarks, as of October 22, 2026.

Read6 min

DeepSeek V4.1 Flash is live on PrimaLabs. Now on PrimaLabs serverless, October 2026.

Engineering · Launches

16 Oct 2026PrimaLabs Engineering

DeepSeek V4.1 Flash is live on PrimaLabs.

DeepSeek V4.1 Flash, with a one-million-token context window, is now on the PrimaLabs serverless API.

Read4 min

Diagram of SciDAC as forward-deployed engineering for science, connecting domain scientists, core algorithms, HPC workflows, and reusable software.

Company · Perspectives

5 Oct 2026Prasanna Balaprakash

SciDAC at 25: A Forward-Deployed Engineering Model for Scientific Discovery

I had the privilege of participating in two discussions marking the 25-year anniversary of the U.S. Department of Energy's Scientific Discovery through Advanced Computing program, better known as SciDAC. I returned in a new role, after years at Argonne and Oak Ridge and now as co-founder and CEO of PrimaLabs, which made the experience especially meaningful.

Read8 min

MiMo-V2.6-Pro. 9.0× faster output. Measured by Artificial Analysis, 24 Sep 2026.

Engineering · Benchmarks

24 Sep 2026PrimaLabs Engineering

MiMo-V2.6-Pro. 9.0× faster output.

MiMo-V2.6-Pro on PrimaLabs serverless: 407 output tokens per second versus Xiaomi’s 45 at 10,000 input tokens. 9.0× faster output, with zero data retention.

Read7 min

The fastest DeepSeek V4 Flash endpoint — measured by Artificial Analysis.

Engineering · Benchmarks

16 Sep 2026PrimaLabs Engineering

The fastest DeepSeek V4 Flash endpoint

Artificial Analysis benchmarked the leading providers serving DeepSeek V4 Flash 0731 on the same harness. Across the 15 providers measured alongside us, PrimaLabs finished first on output speed, end-to-end response time and time to first answer token, at $0.06 per million tokens blended.

Read8 min

PrimaLabs partnering with NVIDIA and U.S. National Laboratories for the DOE Genesis Mission to accelerate scientific AI.

Company · Partnerships

July 2026PrimaLabs Team

PrimaLabs joins NVIDIA and national labs to advance the DOE's Genesis Mission

PrimaLabs is partnering with Savannah River National Laboratory, Oak Ridge National Laboratory, and Pacific Northwest National Laboratory, and NVIDIA to bring GPU-accelerated AI to one of the nation's hardest environmental cleanup challenges.

Read3 min