Koneksia Labs
IDENEnglishLog inSign up free
Test mode — balances and payments are not real money.
z-ai/glm-5.3-flash

GLM 5.3 Flash

Modality
text + image + audio + file → text
Context
1,000,000 tokens
Supported parameters
Temperature, Max tokens
Policy
No PDP providerShow policy details
  • Mangoboost: region US, not retained, not used for training, ZDR — does not qualify

A routing-policy match, not a legal compliance certification.

Providers and prices

Prices per 1M tokens in Rupiah, including FX and markup. Sorted by lowest input price. Metrics from our traffic over the last 24 hours. Observations from our traffic, not a provider SLA.

Providers for this model with price, latency, success, region and policy
ProviderInputOutputTime to first response, p50Successful requestsRegionRetention/PDPTermsAvailability
MangoboostRp 2,574Rp 8,910Not measured yetNot measured yetUS
Not retained
  • Not retained
  • ZDR
  • PDP ✕
Not available yetNot reviewed yetAvailable
USD reference
  • Mangoboost: USD 0.13 / USD 0.45 / 1M tokens

Applied rate Rp 18,000 per USD 1, 18 Nov 2026, 14:30 WIB.

Price per 1,000 words

Indonesian estimate

Input Rp 2.574 · Output Rp 8.91 (Mangoboost)

Initial factor: 1,000 tokens / 1,000 words. An estimate, not a billing unit. Billing is per token.

Integration

ROUTER_BASE_URL = https://api.koneksialabs.ai.id/v1 — Replace with your environment's base URL.

import os
from openai import OpenAI

client = OpenAI(
    base_url=os.environ["ROUTER_BASE_URL"],  # configured origin + /v1
    api_key=os.environ["ROUTER_API_KEY"],
)
response = client.chat.completions.create(
    model="z-ai/glm-5.3-flash",
    messages=[{"role": "user", "content": "Halo!"}],
)
print(response.choices[0].message.content)

Automatic routing across providers

The router picks healthy providers that meet your policy, and moves to another provider on 5xx, 429 or timeout. It falls back to another model only when you configure the models field; there is no auto model.

See advanced routing fields