Error 502: what to do when gemini-3.1-pro-preview fails

A 502 means the gateway received an invalid response from the upstream model service: the request reached the gateway, but the hop to the upstream failed.

502 Bad Gateway means the gateway received an invalid response from the upstream model service. The request reached the gateway; the failure happened on the hop from gateway to upstream, usually due to an upstream restart or a dropped connection.

gemini-3.1-pro-preview is served by Google. Everything on this page — triggers, fixes and measured data — is compiled from the real runtime behaviour of this model at the gateway layer.

At this gateway, the most common trigger is: The upstream model service is unavailable or returned a malformed response. The recommended first action is: Retry once — 502s are usually transient.

Common causes

  • The upstream model service is unavailable or returned a malformed response
  • An upstream node is restarting
  • The gateway-to-upstream connection was interrupted
  • The model was temporarily taken offline

How to fix

  • Retry once — 502s are usually transient
  • Switch to an equivalent model from another vendor
  • If only one model keeps returning 502, its upstream is unhealthy
  • Try again later or contact us

Retry with exponential backoff

The snippet below retries when gemini-3.1-pro-preview returns 502, up to 5 attempts, with an increasing wait plus random jitter so concurrent calls do not retry in lockstep. Read the base URL and API key from environment variables — never hardcode them.

import os, time, random
import requests

BASE  = os.getenv("OPENAI_BASE_URL")   # e.g. https://<your-gateway>/v1
KEY   = os.getenv("OPENAI_API_KEY")
MODEL = 'gemini-3.1-pro-preview'


def chat(messages, retries=5):
    """Retry with exponential backoff + jitter."""
    for i in range(retries):
        try:
            r = requests.post(
                BASE + "/chat/completions",
                headers={"Authorization": "Bearer " + KEY},
                json={"model": MODEL, "messages": messages, "stream": True},
                timeout=60,
            )
            if r.status_code == 429 or r.status_code >= 500:
                time.sleep(min(2 ** i + random.uniform(0, 1), 30))
                continue
            r.raise_for_status()
            return r.json()
        except requests.exceptions.Timeout:
            time.sleep(min(2 ** i + random.uniform(0, 1), 30))
    raise RuntimeError("gave up after " + str(retries) + " retries")


print(chat([{"role": "user", "content": "hello"}]))

Key facts for this model

TPS3738.12
Avg latency26006 ms
Success rate100%
API endpointhttps://api.airai.cc/v1
OpenAI-compatibleOpenAI-compatible
VendorGoogle
Context1M
CapabilitiesReasoning, Tools, Files, Vision, Audio
API formatsopenai, openai-response, openai-response-compact, anthropic, gemini, openai-alpha-search
Billing formulap * 2 + cr * 0.2 + c * 12) : tier("200k_plus", p * 4 + cr * 0.4 + c * 18

FAQ

Does sharing one key across several services make 502 more likely?

It is mainly a quota matter, not a fault in the model itself. With billing p * 2 + cr * 0.2 + c * 12) : tier("200k_plus", p * 4 + cr * 0.4 + c * 18, failed requests are not counted toward usage. Check your balance and rate limits in the console before debugging code.

What information should I give support when reporting an issue?

If it only happens in production, it is usually an environment difference, not the model. The 100% success rate is averaged across nodes, so one weak node drags the whole figure down. This model is served by Google, so upstream status follows the vendor’s own announcements. Reproduce it once in a staging environment with the same request body.

How many retry attempts is reasonable?

You usually do not need to change business code, just the call cadence. Retrying is the most effective first step. Measured success rate is 100%, so the first retry usually carries the highest marginal benefit. With billing p * 2 + cr * 0.2 + c * 12) : tier("200k_plus", p * 4 + cr * 0.4 + c * 18, failed requests are not counted toward usage. Set the retry ceiling to 3–5 attempts and add jitter. Use exponential backoff for retryable errors and return immediately for the rest.

How should I schedule batch jobs when gemini-3.1-pro-preview returns 502?

Lower the concurrency first — most throughput complaints disappear once you do. This model has a 1M context window and comes from Google. Measured throughput is 3738.12, a useful ceiling for concurrency. Start with low concurrency, watch it for a few minutes, then scale up.

Other errors on this model

Other models with the same error

Back to the model page

Data updated: 2026-10-10 18:35

Technical SupportLive Support
Back to Top