LithosAI launches ultra-fast inference with Kimi K3 at 800+ tokens/s
JiaZhihao · x · 2026-08-20
LithosAI announced its first pricing tiers, focusing on ultra-fast agentic inference on standard GPUs. The Kimi K3 model achieves over 800 tokens per second per user with full model quality. The Lithos Engine optimizes the serving system rather than degrading the model, supporting any agent harness like Claude Code or Codex. It offers both hosted API and on-prem deployments, with a public API launch set for September 1.
Related event: LithosAI launches Kimi K3 with 800+ tokens/s inference(3 posts)→
More from Infra
- Phison CEO calls CXMT memory scare a 'ghost story', warns of materials shortage — BenBajarin · 2026-10-16
- Ben Bajarin: The Future of Compute Is Fungible as Hyperscalers Design for Workload Flexibility — BenBajarin · 2026-10-16
- AirLLM runs frontier-size models layer-by-layer: 671B on 12GB VRAM — Aiden_Tech_Ai · 2026-10-16
- NVIDIA Launches cuPotential in CUDA-X to Bring AI Agents to Semiconductor Materials Simulation — tomaszbednarz · 2026-10-16
- Q4-quantized coder model on 8-core CPU builds interactive solar system demo in one shot — IntrepidMindExplorer · 2026-10-16
- TokenSpeed MLA 2 ships for Vera Rubin: up to 3.44× faster Kimi K3 decode — zhyncs42 · 2026-10-16