Parasail first to deploy Positron Titan at scale, claiming 4X better cost per token
JaynitMakwana · x · 2026-11-20
With grid power constrained at many data centers, inference cloud Parasail becomes the first to bring Positron's Titan online at scale (5MW), targeting trillions of tokens/day by late 2027. Titan runs multi-trillion-parameter models on commodity low-power memory and is air-cooled, fitting sites that can't add power-hungry hardware. Positron claims 4X better cost-per-token than next-gen NVIDIA/AMD systems, aimed at agent swarms and physical AI.
More from Infra
- xAI-linked SpaceXAI buys 96 acres in Memphis for $5.5M to expand AI datacenter — chrisgrayson · 2026-11-20
- Ollama CEO: Enterprise customers now ask where their data goes, not where models run — ollama · 2026-11-20
- Surge in expert network inquiries signals growing fear of compute industry's regulatory exposure — mattwbaker · 2026-11-20
- Lawyer runs full DeepSeek-V4.1-Flash on 12x V100 at 157-184 tok/s, beating DGX Spark stacks — TumbleweedNew6515 · 2026-11-20
- US Physical AI Runs on Private CAPEX, Frontier Models, and Energy Flexibility — ingliguori · 2026-11-20
- Together AI webinar Dec 15: step-by-step RL post-training run on Kimi K3.1 — togethercompute · 2026-11-20