How to Use the GPT-6 Astra API: First Call, Effort Levels & Migration from GPT-5.6

pip install openai        # Python
npm install openai        # Node.js

curl https://api.evolink.ai/v1/responses \
  -H "Authorization: Bearer $EVOLINK_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-6-astra",
    "input": "Explain the difference between cache reads and cache writes in one paragraph.",
    "reasoning": {"effort": "medium"}
  }'

from openai import OpenAI

client = OpenAI(
    api_key="your-evolink-api-key",
    base_url="https://api.evolink.ai/v1",
)

response = client.responses.create(
    model="gpt-6-astra",
    input="Explain the difference between cache reads and cache writes in one paragraph.",
    reasoning={"effort": "medium"},
)

print(response.output_text)
print(response.model, response.usage)

import OpenAI from "openai";

const client = new OpenAI({
  apiKey: "your-evolink-api-key",
  baseURL: "https://api.evolink.ai/v1",
});

const response = await client.responses.create({
  model: "gpt-6-astra",
  input: "Explain the difference between cache reads and cache writes in one paragraph.",
  reasoning: { effort: "medium" },
});

console.log(response.output_text);
console.log(response.model, response.usage);

response = client.chat.completions.create(
    model="gpt-6-astra",
    messages=[{"role": "user", "content": "Summarize this changelog in three bullets: ..."}],
    reasoning_effort="low",
)
print(response.choices[0].message.content)

response = client.responses.create(
    model="gpt-6-astra",
    input=[{"role": "user", "content": "Refactor this module and explain each change."}],
    reasoning={"effort": "high"},
    max_output_tokens=8000,
)

tools = [{
    "type": "function",
    "name": "get_build_status",
    "description": "Return the status of the latest CI build for a branch.",
    "parameters": {
        "type": "object",
        "properties": {"branch": {"type": "string"}},
        "required": ["branch"],
    },
}]

first = client.responses.create(
    model="gpt-6-astra",
    input=[{"role": "user", "content": "Is the main branch green? If not, summarize the failure."}],
    tools=tools,
    reasoning={"effort": "medium"},
)

calls = [item for item in first.output if item.type == "function_call"]
outputs = []
for call in calls:
    # run your tool here
    outputs.append({
        "type": "function_call_output",
        "call_id": call.call_id,
        "output": '{"status": "failed", "step": "unit-tests", "log_url": "https://ci.example/123"}',
    })

second = client.responses.create(
    model="gpt-6-astra",
    input=[{"role": "user", "content": "Is the main branch green? If not, summarize the failure."}]
          + list(first.output) + outputs,
    tools=tools,
    reasoning={"effort": "medium"},
)
print(second.output_text)

response = client.responses.create(
    model="gpt-6-astra",
    input="Extract the model ID, context window, and output limit from this text: ...",
    text={
        "format": {
            "type": "json_schema",
            "name": "model_spec",
            "schema": {
                "type": "object",
                "properties": {
                    "model_id": {"type": "string"},
                    "context_tokens": {"type": "integer"},
                    "max_output_tokens": {"type": "integer"},
                },
                "required": ["model_id", "context_tokens", "max_output_tokens"],
                "additionalProperties": False,
            },
            "strict": True,
        }
    },
    reasoning={"effort": "low"},
)
print(response.output_text)

 response = client.chat.completions.create(
-    model="gpt-5.6-sol",
+    model="gpt-6-astra",
     messages=messages,
-    temperature=0.2,
-    reasoning_effort="none",
+    reasoning_effort="low",
 )

def call_with_fallback(**kwargs):
    for model in ("gpt-6-astra", "gpt-5.6-sol"):
        try:
            return client.responses.create(model=model, **kwargs)
        except Exception as err:  # narrow this to 429/5xx in production
            last = err
    raise last