GLM 5.2
- Khusus model glm-5.2
Pick a package, pay once, and get an API key instantly. One key for DeepSeek, GLM, Qwen, Kimi and more.
Starting from Rp18,000
Token quota counts input + output. Prefer to pay only for what you use? See pay-as-you-go rates
Create an account in under a minute.
Pay via QRIS or TOKS balance. Processed automatically.
Get a key instantly — works with any OpenAI-compatible app.
Top up TOKS balance and pay only for tokens used. Balance never expires.
| Model | Provider | Context | Input / 1M tokens | Output / 1M tokens | Status |
|---|---|---|---|---|---|
GLM-5.2 glm-5.2 | GLM | 1M | Rp2,000 2 TOKS | Rp4,400 4,4 TOKS | Active |
DeepSeek V4 Flash 0731 deepseek-v4-flash-0731 | DeepSeek | 1M | Rp1,200 1,2 TOKS | Rp2,400 2,4 TOKS | Active |
DeepSeek V4.1 Flash deepseek-v4.1-flash | DeepSeek | 1M | Rp1,200 1,2 TOKS | Rp2,400 2,4 TOKS | Active |
Kimi K3 kimi-k3 | Moonshot AI | 256K | Rp4,000 4 TOKS | Rp5,000 5 TOKS | Active |
DeepSeek V4 Pro 0813 deepseek-v4-pro-0813 | DeepSeek | 1M | Rp2,000 2 TOKS | Rp4,000 4 TOKS | Active |
Qwen 8 Max qwen3.8-max | Alibaba | 200K | Rp3,000 3 TOKS | Rp6,000 6 TOKS | Active |
GLM 5.3 Flash glm-5.3-flash | GLM | 1000K | Rp1,400 1,4 TOKS | Rp3,000 3 TOKS | Active |
GLM 5.3 glm-5.3 | GLM | 1M | Rp2,500 2,5 TOKS | Rp3,500 3,5 TOKS | Active |
Everything you need to ship reliable AI features — from routing to billing — behind one predictable API.
One endpoint for many frontier AI models. Switch models with a single parameter — no SDK changes.
Routing optimized for low latency. Smart failover keeps requests fast and resilient.
Works with your existing SDK. Just change the base URL — no rewrite, no lock-in.
Enterprise-grade API key management, per-key rate limits, and encrypted credentials. Your data stays yours.
8 production-ready models from leading providers — all behind one API key, billed per token.
Already using the OpenAI SDK? Just point your client to our base URL — done.
import OpenAI from "openai";
const client = new OpenAI({
baseURL: "https://9inference.cloud/v1",
apiKey: process.env.NINEINFERENCE_API_KEY,
});
const res = await client.chat.completions.create({
model: "glm-5.2",
messages: [
{ role: "user", content: "Hello!" }
],
});
console.log(res.choices[0].message.content);A token package is a one-time purchase that gives you a fixed number of tokens (input + output) for a set active period. You get a dedicated API key right after payment — no subscription, no auto-renewal.
Pick a token package if you want a fixed, cheaper price for a known amount of usage. Pick pay-as-you-go (top up TOKS balance) if your usage is irregular — you only pay for the tokens you actually use, and balance does not expire.
9inference is an affordable AI model API gateway from Indonesia. One OpenAI-compatible endpoint gives you access to DeepSeek, GLM, Qwen, Kimi, MiniMax and more — transparent pricing in rupiah, no subscription.
Yes. Point your existing OpenAI client to the 9inference base URL, use your 9inference API key, and keep your code unchanged. It works with the official openai SDK, LangChain, and any OpenAI-compatible tool.
Via QRIS (all Indonesian e-wallets and mobile banking) or with your TOKS balance. 1 TOKS = Rp1,000. No credit card needed.
DeepSeek, GLM, Alibaba Qwen, Moonshot AI Kimi, MiniMax and more. Each package card shows which models it covers.
Pick a token package, pay via QRIS, and your API key is ready in seconds.