cd /news/artificial-intelligence/breaking-openai-value-can-now-return… · home › topics › artificial-intelligence › article
[ARTICLE · art-163496] src=driftwatch.et ↗ pub= topic=artificial-intelligence verified=true sentiment=· neutral

OpenAI: value can now return "max_output_tokens"

OpenAI added "max_output_tokens" as a new value on its SessionTurnErrorCodeResource, an enum reachable from 6 OpenAI operations, according to the company's published API specification. The change is additive on the producer side but breaks exhaustive switch statements, strict parsers and generated client enums that do not handle the new value. Code reading SessionTurnErrorCodeResource objects now must account for "max_output_tokens" alongside the 23 existing values, including "context_length_exceeded" and "rate_limit_exceeded".

read1 min views3 publishedNov 3, 2026

What changed #

Producer-side tools call this additive. It breaks exhaustive switches, strict parsers and generated client enums.

Affects this value on SessionTurnErrorCodeResource reachable from 6 operations.

New value: max_output_tokens.

can now return "max_output_tokens"

Who is affected #

Code that reads OpenAI's SessionTurnErrorCodeResource objects, used by 6 OpenAI operations.

Before and after #

Where Before After
SessionTurnErrorCodeResource "active_turn_not_steerable", "authentication_error", "billing_not_active", "connection_failed", "context_length_exceeded", "credit_balance_exhausted", "cyber_policy", "executor_version_incompatible", "flex_unavailable", "internal_error", "invalid_request", "misalignment_policy_violation", "organization_spend_limit_exceeded", "organization_usage_limit_exceeded", "project_spend_limit_exceeded", "rate_limit_exceeded", "request_timeout", "resource_not_found", "sandbox_error", "server_error", "server_overloaded", "session_budget_exceeded", "usage_limit_exceeded" "active_turn_not_steerable", "authentication_error", "billing_not_active", "connection_failed", "context_length_exceeded", "credit_balance_exhausted", "cyber_policy", "executor_version_incompatible", "flex_unavailable", "internal_error", "invalid_request", "max_output_tokens", "misalignment_policy_violation", "organization_spend_limit_exceeded", "organization_usage_limit_exceeded", "project_spend_limit_exceeded", "rate_limit_exceeded", "request_timeout", "resource_not_found", "sandbox_error", "server_error", "server_overloaded", "session_budget_exceeded", "usage_limit_exceeded"

Code that breaks #

switch (response) {
  // existing branches …
  // no branch for "max_output_tokens"
  default:
    throw new Error("unexpected value");
}

Based on OpenAI’s published API specification. How severity is decided.

── more in #artificial-intelligence 4 stories · sorted by recency
── more on @openai 3 stories trending now
sponsored brought to you by zahid.host 4,200+ EU-deployed projects
reading about agents? ship yours in a single git push.

Run your AI side-project on zahid.host

EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain — perfect for shipping the agent you just read about.

$git push zahid main
→ Live at https://your-agent.zahid.host ✓
Get free account → Pricing
from €0/mo · no card required
LIVE [news/breaking-openai-valu…] indexed:0 read:1min 2026-11-03 · —