Home › Blog › GLM API Guide

GLM API Guide 2026: GLM-5.2 Pricing, Vision Input & OpenAI Compatibility

Zhipu's GLM-5.2 is the strongest Chinese-language model with vision input — but official pricing (¥8 in / ¥28 out per 1M tokens) makes it a premium purchase. Bridge serves it at $0.40/$1.20 through your existing OpenAI SDK.

GLM-5.2 pricing: official vs Bridge

When to choose GLM-5.2

• Chinese-language understanding — customer support, content moderation and localization pipelines where GLM leads.
• Vision input — image understanding via the same chat/completions endpoint, compatible with OpenAI's image_url message format.
• Multimodal agents — screenshots, forms and UI analysis combined with strong reasoning.

Vision example (OpenAI SDK, unchanged)

from openai import OpenAI

client = OpenAI(
    base_url="https://bridge.computeindex.net/v1",
    api_key="sk-bridge-YOUR_KEY",
)

completion = client.chat.completions.create(
    model="glm-5.2",
    messages=[{
        "role": "user",
        "content": [
            {"type": "text", "text": "What does this screenshot show?"},
            {"type": "image_url", "image_url": {"url": "https://example.com/shot.png"}},
        ],
    }],
)
print(completion.choices[0].message.content)

Cache hits cut GLM bills by 90%

Like DeepSeek, GLM prices cached input at a steep discount — on Bridge, cache hits cost $0.04 per 1M tokens. Reusing a system prompt across thousands of turns makes the effective cost of conversation dramatically lower; the usage receipt on every request shows exactly how many tokens were cached.

FAQ

Is Bridge's GLM-5.2 the same model as Zhipu's official? Yes — same model served through Bridge's OpenAI-compatible gateway, with per-token metering and cached-input billing.

Can I use GLM from an existing OpenAI integration? Yes. Change base_url and api_key; messages, streaming and error codes follow the OpenAI dialect.