LithosAI Announces Pricing Tiers for Kimi K3
The firm launches pricing tiers ahead of its September 1 API release.
Zhihao Jia posted that LithosAI has released its first pricing tiers with early-access rates before the September 1 API launch. He stated that Kimi K3 now runs live at 800+ tokens per second per user on standard GPUs while preserving full model quality. The LithosAI site claims the system delivers 1,000 tokens per second on a 1T parameter model and runs 3.4 to 5.7 times faster than major providers on a standard 8xB200 GPU node. Tim Dettmers and Tianqi Chen both reposted the announcement.
Combined views
12.3K
4 posts, first seen 5d ago
Reactions from ranked influencers
4 posts
🚀We’ve been pushing agentic inference toward the physical limits of the hardware. Announcing LithosAI’s first pricing tiers, with early-access pricing ahead of the September 1 API launch. Kimi K3 is live now at 800+ tokens/sec/user on standard GPUs, with full model quality.…
🚀Tired of slow agents? We’re launching LithosAI’s ultra-fast Kimi K3 pricing tiers. If you’re building agents where latency matters, we can give you the fastest inference available across models, through a hosted API or on-prem. Kimi K3 chat is live today at 800+…
🚀We’ve been pushing agentic inference toward the physical limits of the hardware. Announcing LithosAI’s first pricing tiers, with early-access pricing ahead of the September 1 API launch. Kimi K3 is live now at 800+ tokens/sec/user on standard GPUs, with full model quality.…
🚀Tired of slow agents? We’re launching LithosAI’s ultra-fast Kimi K3 pricing tiers. If you’re building agents where latency matters, we can give you the fastest inference available across models, through a hosted API or on-prem. Kimi K3 chat is live today at 800+…
🚀We’ve been pushing agentic inference toward the physical limits of the hardware. Announcing LithosAI’s first pricing tiers, with early-access pricing ahead of the September 1 API launch. Kimi K3 is live now at 800+ tokens/sec/user on standard GPUs, with full model quality.…
🚀Tired of slow agents? We’re launching LithosAI’s ultra-fast Kimi K3 pricing tiers. If you’re building agents where latency matters, we can give you the fastest inference available across models, through a hosted API or on-prem. Kimi K3 chat is live today at 800+…
🚀We’ve been pushing agentic inference toward the physical limits of the hardware. Announcing LithosAI’s first pricing tiers, with early-access pricing ahead of the September 1 API launch. Kimi K3 is live now at 800+ tokens/sec/user on standard GPUs, with full model quality.…
🚀Tired of slow agents? We’re launching LithosAI’s ultra-fast Kimi K3 pricing tiers. If you’re building agents where latency matters, we can give you the fastest inference available across models, through a hosted API or on-prem. Kimi K3 chat is live today at 800+…