JSON Mode and Beyond

In this tutorial, you’ll learn how to:

Guarantee structured output using our chat completions API
This can be done using JSON Schema / Pydantic models, (schemaless) JSON Mode, and multiple choice.

Sentiment Analysis using Multiple Choice

One of the simplest forms of structured output is multiple choice. The schema below pins the classification to a single value — the response is always {"sentiment": "positive"} or {"sentiment": "negative"}.

Make sure you have set the TELNYX_API_KEY environment variable

zai-org/GLM-5.3-Flash is a reasoning model. Your structured output still arrives in choices[0].message.content — the model’s chain-of-thought is returned separately in choices[0].message.reasoning_content, so it never pollutes the JSON you parse. To surface the reasoning, read it alongside content:

message = chat_completion.choices[0].message
if getattr(message, "reasoning_content", None):
    print("reasoning:", message.reasoning_content)
print(message.content)  # the structured output

import os
from openai import OpenAI

client = OpenAI(
    api_key=os.getenv("TELNYX_API_KEY"),
    base_url="https://api.telnyx.com/v2/ai/openai",
)

chat_completion = client.chat.completions.create(
    model="zai-org/GLM-5.3-Flash",
    reasoning_effort="high",
    messages=[
        {"role": "system", "content": "Classify the sentiment of this review as positive or negative."},
        {"role": "user", "content": "The staff went above and beyond! I had a great stay."}
    ],
    temperature=0.0,
    response_format={
        "type": "json_schema",
        "json_schema": {
            "name": "sentiment",
            "schema": {
                "type": "object",
                "properties": {
                    "sentiment": {
                        "type": "string",
                        "enum": ["positive", "negative"]
                    }
                },
                "required": ["sentiment"],
                "additionalProperties": False
            }
        }
    }
)
print(chat_completion.choices[0].message.content)

Since the schema’s sentiment property is an enum with exactly two values, the model can only answer with one of positive or negative:

{"sentiment": "positive"}

Constraining the full response shape with `json_schema`

Now let’s say we want to capture an explanation for the classification as well. The response_format parameter with "type": "json_schema" guarantees the response parses as JSON matching your schema, and "additionalProperties": false on an object schema makes the response carry exactly the declared properties — without it, JSON Schema permits extra keys, so a schema listing sentiment and explanation doesn’t by itself forbid a third key:

import os
from openai import OpenAI

client = OpenAI(
    api_key=os.getenv("TELNYX_API_KEY"),
    base_url="https://api.telnyx.com/v2/ai/openai",
)

chat_completion = client.chat.completions.create(
    model="zai-org/GLM-5.3-Flash",
    reasoning_effort="high",
    messages=[
        {"role": "system", "content": "First describe the sentiment of the following review and then use that to classify the sentiment as positive or negative."},
        {"role": "user", "content": "The staff went above and beyond! I had a great stay."}
    ],
    temperature=0.0,
    response_format={
        "type": "json_schema",
        "json_schema": {
            "name": "sentiment_analysis",
            "schema": {
                "type": "object",
                "properties": {
                    "sentiment": {
                        "type": "string",
                        "enum": ["positive", "negative"]
                    },
                    "explanation": {"type": "string"}
                },
                "required": ["sentiment", "explanation"],
                "additionalProperties": False
            }
        }
    }
)
print(chat_completion.choices[0].message.content)

This will ensure a JSON response with the same schema as the following

{
  "sentiment": "positive",
  "explanation": "The review mentions that the staff went above and beyond and the user had a great stay, which indicates a positive sentiment."
}

The guided_json, guided_choice, and guided_regex request fields are accepted for compatibility with the vLLM/SGLang serving stack, but they are not enforced for Telnyx-hosted models today: a request carrying them is served as a plain chat completion, so the model may return free text that ignores the schema or choice list. Use response_format with "type": "json_schema" for guaranteed structured output on Telnyx-hosted models.

Simplifying Schema Generation using Pydantic

The above is helpful to see the raw JSON Schema specification being sent via API. However, for practical purposes, using Pydantic models can help simplify generating the specs. The following is functionally equivalent to the previous example.

import os
from enum import Enum
from openai import OpenAI
from pydantic import BaseModel, ConfigDict

client = OpenAI(
    api_key=os.getenv("TELNYX_API_KEY"),
    base_url="https://api.telnyx.com/v2/ai/openai",
)

class Sentiment(str, Enum):
    positive = 'positive'
    negative = 'negative'

class SentimentAnalysis(BaseModel):
    model_config = ConfigDict(extra='forbid')
    explanation: str
    sentiment: Sentiment

chat_completion = client.chat.completions.create(
    model="zai-org/GLM-5.3-Flash",
    reasoning_effort="high",
    messages=[
        {"role": "system", "content": "First describe the sentiment of the following review and then use that to classify the sentiment as positive or negative."},
        {"role": "user", "content": "The staff went above and beyond! I had a great stay."}
    ],
    temperature=0.0,
    response_format={
        "type": "json_schema",
        "json_schema": {
            "name": "SentimentAnalysis",
            "schema": SentimentAnalysis.model_json_schema()
        }
    }
)
print(chat_completion.choices[0].message.content)

Schema-less JSON Mode

We also support the schema-less JSON mode provided by OpenAI

import os
from openai import OpenAI


client = OpenAI(
    api_key=os.getenv("TELNYX_API_KEY"),
    base_url="https://api.telnyx.com/v2/ai/openai",
)

chat_completion = client.chat.completions.create(
    model="zai-org/GLM-5.3-Flash",
    reasoning_effort="high",
    messages=[
        {"role": "system", "content": "First describe the sentiment of the following review and then use that to classify the sentiment as positive or negative. Please respond using a JSON object."},
        {"role": "user", "content": "The staff went above and beyond! I had a great stay."}
    ],
    temperature=0.0,
    response_format={"type": "json_object"}
)
print(chat_completion.choices[0].message.content)

Inference

Data Residency

Integrations

Tutorials

API Reference

For AI Agents

JSON Mode and Beyond

Sentiment Analysis using Multiple Choice

Constraining the full response shape with `json_schema`

Simplifying Schema Generation using Pydantic

Schema-less JSON Mode

​Sentiment Analysis using Multiple Choice

​Constraining the full response shape with json_schema

​Simplifying Schema Generation using Pydantic

​Schema-less JSON Mode

Sentiment Analysis using Multiple Choice

Constraining the full response shape with `json_schema`

Simplifying Schema Generation using Pydantic

Schema-less JSON Mode