$ docker run -it   --name "spec_decoding_structured_output"   --shm-size 256g   --gpus all  -v "/tmp:/tmp"  --ipc=host   --network=host   --privileged --entrypoint=bash   lmsysorg/sglang:latest

$ docker exec -it spec_decoding_structured_output bash

$ python3 -m sglang.launch_server --port=7080 --model=/tmp/openai/gpt-oss-120b --trust-remote-code --tp=8 --max-queued-requests=256 --tool-call-parser=gpt-oss --reasoning-parser=gpt-oss --speculative-algorithm=EAGLE3 --speculative-draft-model-path=/tmp/lmsys/EAGLE3-gpt-oss-120b-bf16 --speculative-num-steps=5 --speculative-eagle-topk=4 --speculative-num-draft-tokens=8 --enable-metrics --log-requests-level=1 --enable-request-time-stats-logging 

$ cat json_schema.py 
import openai
import os

from pydantic import BaseModel, Field
from sglang.utils import wait_for_server, print_highlight, terminate_process

port=7080
client = openai.Client(base_url=f"http://127.0.0.1:{port}/v1", api_key="None")

# Define the schema using Pydantic
class CapitalInfo(BaseModel):
    name: str = Field(..., pattern=r"^\w+$", description="Name of the capital city")
    population: int = Field(..., description="Population of the capital city")


response = client.chat.completions.create(
    model="openai/gpt-oss-120b",
    messages=[
        {
            "role": "user",
            "content": "Please generate the information of the capital of France in the JSON format.",
        },
    ],
    temperature=0,
    max_tokens=128,
    response_format={
        "type": "json_schema",
        "json_schema": {
            "name": "foo",
            # convert the pydantic model to json schema
            "schema": CapitalInfo.model_json_schema(),
        },
    },
)

print(f"===Pydantic json response===\n{response}")
print("\n\n")
# response_content = response.choices[0].message.content
# validate the JSON response by the pydantic model
# capital_info = CapitalInfo.model_validate_json(response_content)
# print_highlight(f"Validated response: {capital_info.model_dump_json()}")

import json

json_schema = json.dumps(
    {
        "type": "object",
        "properties": {
            "name": {"type": "string", "pattern": "^[\\w]+$"},
            "population": {"type": "integer"},
        },
        "required": ["name", "population"],
    }
)

response = client.chat.completions.create(
    model="openai/gpt-oss-120b",
    messages=[
        {
            "role": "user",
            "content": "Give me the information of the capital of France in the JSON format.",
        },
    ],
    temperature=0,
    max_tokens=128,
    response_format={
        "type": "json_schema",
        "json_schema": {"name": "foo", "schema": json.loads(json_schema)},
    },
)
print(f"===JSON Schema Directly===:\n{response}")
print("\n\n")

ebnf_grammar = """
root ::= city | description
city ::= "London" | "Paris" | "Berlin" | "Rome"
description ::= city " is " status
status ::= "the capital of " country
country ::= "England" | "France" | "Germany" | "Italy"
"""

response = client.chat.completions.create(
    model="openai/gpt-oss-120b",
    messages=[
        {"role": "system", "content": "You are a helpful geography bot."},
        {
            "role": "user",
            "content": "Give me the information of the capital of France.",
        },
    ],
    temperature=0,
    max_tokens=32,
    extra_body={"ebnf": ebnf_grammar},
)

print(f"===EBNF===\n{response}")


$ python json_schema.py 
===Pydantic json response===
ChatCompletion(id='a6b6981bb93b44febb99c098eaeb0383', choices=[Choice(finish_reason='length', index=0, logprobs=None, message=ChatCompletionMessage(content=None, refusal=None, role='assistant', annotations=None, audio=None, function_call=None, tool_calls=None, reasoning_content=None), matched_stop=None)], created=1762913414, model='openai/gpt-oss-120b', object='chat.completion', service_tier=None, system_fingerprint=None, usage=CompletionUsage(completion_tokens=128, prompt_tokens=81, total_tokens=209, completion_tokens_details=None, prompt_tokens_details=None, reasoning_tokens=0), metadata={'weight_version': 'default'})



===JSON Schema Directly===:
ChatCompletion(id='fb9792fc660a4b5180d7cca9d6ca1353', choices=[Choice(finish_reason='length', index=0, logprobs=None, message=ChatCompletionMessage(content=None, refusal=None, role='assistant', annotations=None, audio=None, function_call=None, tool_calls=None, reasoning_content=None), matched_stop=None)], created=1762913415, model='openai/gpt-oss-120b', object='chat.completion', service_tier=None, system_fingerprint=None, usage=CompletionUsage(completion_tokens=128, prompt_tokens=81, total_tokens=209, completion_tokens_details=None, prompt_tokens_details=None, reasoning_tokens=0), metadata={'weight_version': 'default'})



===EBNF===
ChatCompletion(id='f2199a2eaf684706bbe6f3ad4bdb17bd', choices=[Choice(finish_reason='length', index=0, logprobs=None, message=ChatCompletionMessage(content=None, refusal=None, role='assistant', annotations=None, audio=None, function_call=None, tool_calls=None, reasoning_content=None), matched_stop=None)], created=1762913415, model='openai/gpt-oss-120b', object='chat.completion', service_tier=None, system_fingerprint=None, usage=CompletionUsage(completion_tokens=32, prompt_tokens=91, total_tokens=123, completion_tokens_details=None, prompt_tokens_details=None, reasoning_tokens=0), metadata={'weight_version': 'default'})