Every call free, nothing drawn from your balance — until September 30, 2026. And the $100 stays ordinary credit, spendable across the other 1050+ models. Nothing to set up: same key, same endpoint.
The tiers are exclusive on purpose: at $100 the unlock is the reward, so there is no bonus on top of unmetered access.
Free calls until the offer closes. Your $100 stays put — spend it on any of the other 1050+ models whenever you like. Unmetered, not unthrottled: 600 requests a minute.
Put in $50, get $150 to spend on Flash. The bonus is capped at $100 per account, so $50 is the sweet spot. Expires 90 days after the top-up, spendable only on GLM-5.3 Flash.
GLM-5.3 Flash approaches Claude Opus 4.8 on coding and agentic benchmarks, natively multimodal, with a 1,310,720-token context window. Normally $0.15/M input, $0.50/M output.
If your code calls the OpenAI SDK, it already calls AIgateway — swap the base URL and model slug and you are on Flash. Works in Cursor, Cline, Aider, Continue, LangChain, the Vercel AI SDK, and everything else that speaks OpenAI.
from openai import OpenAI client = OpenAI( base_url="https://api.aigateway.sh/v1", api_key="sk-aig-...", ) r = client.chat.completions.create( model="zai-org/glm-5.3-flash", # unmetered for you messages=[{"role": "user", "content": "Ship it."}], )
Unlimited means unmetered, not unthrottled — 600 requests a minute. Above that you get 429 with Retry-After.
It takes one $100 payment; smaller top-ups don't add up to it. Lasts while your top-ups net of refunds stay at or above that amount — refunding takes it back.
Both offers end September 30, 2026 and won't run again. After that, Flash bills at its normal rate.
Your credit doesn't expire with the offer — it stays yours, spendable on any model. The sub-$100 bonus credit expires 90 days after the top-up and is capped at $100 per account.