Less than 11 remaining
Private Model API: GLM-5.3-Flash, up to 1M Context
$25/ month
6 members
Private OpenAI-compatible GLM API: long context, your own stream, no training
Choose a plan (monthly + tax until you cancel; charged only when you confirm at checkout):
Private, OpenAI-compatible API access to GLM-5.3-Flash (NVIDIA FP4 checkpoint), served from one US workstation.
- Private: prompts and outputs are never used for training or passed to a third-party AI provider; our gateway records usage metadata only (token counts, timing), not your prompts or outputs
- Context: 128K (Solo), 256K (Pro), 512K or 1M (Reserved) tokens per request
- Each API key has its own stream limit (1 on Solo, 2 on Pro); capacity is shared with other customers and queues by tier when busy
- Monthly token allowances: 500M on Solo, 2B on Pro; Reserved: no fixed monthly token cap (fair use)
- Tailscale access with your own API key (no public endpoint)
- Spare capacity on owner-run hardware: owner and client workloads take precedence; when busy, requests may queue or get a 503 with Retry-After; per-key limits return 429. No uptime SLA
Licensed by NVIDIA Corporation under the NVIDIA Open Model Agreement; base model GLM-5.3-Flash is MIT-licensed © 2026 Z.AI Co., Ltd. Not affiliated with or endorsed by NVIDIA. US customers only. Sold by SlowGlow AI LLC (NC). Terms, AUP and Privacy Notice apply.