VESSL AI

Liquid AI Infra & Persistent GPU Cloud — A100 / H100 / B200 / B300 / GB200

• 🖥️ On-demand GPU workspaces & multi-node clusters with InfiniBand
• ⚡ OpenAI- and Anthropic-compatible inference endpoints for open and fine-tuned models
• 💾 Storage tiers built for large model artifacts and datasets

On the Hub: vessl/Kimi-K3-W4AFP8 — Kimi K3 quantized to W4A-FP8 for efficient sglang inference.

🌐 vessl.ai · 💻 GitHub · 📝 Blog · 🐦 X · 💼 LinkedIn · ▶️ YouTube