glm

Call the GLM API (glm-5.2, glm-5.1, glm-5-turbo, glm-5, glm-4.7, glm-4.6, glm-4.5, glm-4.5-air) through RunAPI using the official OpenAI SDK or compatible clients. Use when the user asks for GLM chat, streaming completions, Anthropic or Gemini protocol compatibility, or when they want to point an existing OpenAI SDK setup at RunAPI as the base URL.

GLM on RunAPI

Use the official OpenAI SDK or any OpenAI-compatible HTTP client and switch the base URL to https://runapi.ai/v1. The primary endpoint is Chat Completions (POST /v1/chat/completions).

Setup

dotenv

OPENAI_API_KEY=YOUR_RUNAPI_TOKEN
OPENAI_BASE_URL=https://runapi.ai/v1

Get a RunAPI API Key at https://runapi.ai/api_keys.

Core recipe - Chat Completions

python

from openai import OpenAI

client = OpenAI(
    api_key="YOUR_RUNAPI_TOKEN",
    base_url="https://runapi.ai/v1",
)

response = client.chat.completions.create(
    model="glm-5.2",
    messages=[{"role": "user", "content": "Summarize this design review."}],
)
print(response.choices[0].message.content)
print(response.usage)

typescript

import OpenAI from "openai";

const client = new OpenAI({
  apiKey: "YOUR_RUNAPI_TOKEN",
  baseURL: "https://runapi.ai/v1",
});

const response = await client.chat.completions.create({
  model: "glm-5.2",
  messages: [{ role: "user", content: "Summarize this design review." }],
});

bash

curl -X POST "https://runapi.ai/v1/chat/completions" \
  -H "Authorization: Bearer YOUR_RUNAPI_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "glm-5.2",
    "messages": [{"role": "user", "content": "Hello, GLM!"}]
  }'

Streaming

python

stream = client.chat.completions.create(
    model="glm-5-turbo",
    messages=[{"role": "user", "content": "Write a short implementation plan."}],
    stream=True,
)
for chunk in stream:
    delta = chunk.choices[0].delta.content
    if delta:
        print(delta, end="", flush=True)

Streaming returns partial output as it is generated. Use it for long responses.

Protocol compatibility

GLM models are also available through RunAPI's Anthropic-compatible Messages and Gemini-compatible contents interfaces. Use these protocol paths when an existing agent runtime requires their request shapes:

bash

curl -X POST "https://runapi.ai/v1/messages" \
  -H "x-api-key: YOUR_RUNAPI_TOKEN" \
  -H "anthropic-version: 2023-06-01" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "glm-4.6",
    "max_tokens": 1024,
    "messages": [{"role": "user", "content": "Draft a concise answer."}]
  }'

bash

curl -X POST \
  "https://runapi.ai/v1beta/models/glm-4.6:streamGenerateContent" \
  -H "x-goog-api-key: YOUR_RUNAPI_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"contents":[{"role":"user","parts":[{"text":"Hello!"}]}]}'

For new app code, prefer the OpenAI-compatible setup.

List models

bash

curl https://runapi.ai/v1/models \
  -H "Authorization: Bearer YOUR_RUNAPI_TOKEN"

Supported models

Model ID	Use when
glm-5.2	Current flagship text workloads
glm-5.1	Latest GLM chat workloads
glm-5-turbo	Faster GLM chat
glm-5	General GLM 5 requests
glm-4.7	GLM 4.7 compatibility
glm-4.6	Stable GLM 4.6 requests
glm-4.5	GLM 4.5 compatibility
glm-4.5-air	Lightweight GLM 4.5 requests

GLM-5.2 capability contract

Capability	Current behavior
Text history, sync, streaming	Supported across Chat, Responses, Messages, and Gemini-compatible requests
Default thinking	Disabled for consistent cross-protocol output
Reasoning controls/content, function tools/history, structured output	Reject with unsupported_model_capability; do not retry without the field silently
Hosted tools, explicit cache controls, stateful continuation, protocol-specific controls	Reject; RunAPI does not currently expose the required protocol surface
Cache/reasoning Usage	Preserve returned fields and never infer omitted details; Messages has no caller-visible reasoning-token breakdown
Context window	Keep estimated input at or below 1000000 tokens; full-window execution remains readiness-gated
Maximum output	Keep the protocol-specific value at or below 131072; larger values return model_limit_exceeded before request execution
Multimodal input	Reject with unsupported_model_capability

Do not present the model's 1M context envelope as a RunAPI execution guarantee. Full-window execution remains subject to published model readiness.

References

• Model overview, pricing, and rate limits: https://runapi.ai/models/glm.md
• Provider comparison: https://runapi.ai/providers/z-ai.md
• Catalog: https://runapi.ai/models.md

Agent rules

• Keep API keys in OPENAI_API_KEY, RUNAPI_TOKEN, or a secret manager; never inline them in commits or shell history.
• Default new integrations to the OpenAI-compatible client at https://runapi.ai/v1.
• Keep glm-5.2 requests text-only. Use only text history until the capability table marks an advanced feature ready.
• Do not remove a rejected field and retry unless the user explicitly accepts the semantic change.
• Treat cache and reasoning Usage as response facts. Never infer missing token details or assume a cache discount.
• Use streaming for responses longer than a few hundred tokens.
• For pricing, rate-limit, and commercial-usage answers, link to https://runapi.ai/models/glm.md rather than copying values.