For AI agents: markdown of this page — /docs-content-en/changelog/ai-chat.md documentation index — /llms.txt

Change history: Chat Completions

← Changelog. Entries: 48.

2026-09-28 BC-0928-7: the agent_turn_loop_detected refusal for a looping agent turn is switched on
2026-09-28 FIX-0928-26: a chat response without tool calls is no longer marked as a tool call
2026-09-26 FIX-0926-3: looping model reasoning is cut off with the reasoning_repetition code
2026-09-26 FIX-0926-4: the streamed chat response always ends with a finish reason
2026-09-25 FIX-0925-11: BitrixGPT 5.5 accepts json_schema when the model supports structured outputs
2026-09-24 NEW-0924-10: new refusal code agent_turn_loop_detected for a looping agent turn
2026-09-22 NEW-0922-2: two new refusal codes for agent runtime keys
2026-09-20 FIX-0920-8: the previous turn's reasoning reaches the model again
2026-09-18 FIX-0918-7: a function-call turn no longer duplicates the reasoning into content
2026-09-16 NEW-0916-8: Cowork off-peak and quota-relief keys now always arrive
2026-09-15 FIX-0915-6: the frozen-account refusal no longer promises the backup model is paid from the wallet
2026-09-14 BC-0914-19: the chat model field no longer matches models by substring
2026-09-14 NEW-0914-22: reasoning from earlier responses in the chat completion history
2026-09-14 FIX-0914-24: the default reasoning step is sent to the model explicitly
2026-09-10 BC-0910-1: a network failure on the way to the provider arrives under its own code ai_provider_network
2026-09-10 FIX-0910-2: AI provider errors now carry the pause from the provider's own header, in one shape for every stream
2026-09-10 FIX-0910-7: responses of a model hosted by a third-party provider name the public model id
2026-09-10 BC-0910-11: a custom AI provider no longer follows redirects
2026-09-09 FIX-0909-11: model reasoning for Cowork no longer appears in final text
2026-09-05 FIX-0905-1: chat with a Cowork/Code key works while the product is disabled
2026-09-03 NEW-0903-2: reasoning control in chat completions
2026-09-01 FIX-0901-9: file uploads and AI requests now answer 429 under memory pressure
2026-08-22 FIX-0822-13: The inactive-subscription refusal now points at the subscription page
2026-08-20 FIX-0820-8: AI provider refusal: a readable message instead of relayed prose, and one error shape for streaming and non-streaming
2026-08-16 NEW-0816-3: new 402 company_budget_exhausted rejection on calls that spend credits
2026-08-14 FIX-0814-4: leading service markers are no longer included in model content
2026-08-11 NEW-0811-4: off-peak hours are visible in the Cowork/Code subscription and in the exhausted-quota refusal
2026-08-11 FIX-0811-23: the ai_congested retry pause now scales with the configured base and is capped
2026-08-11 BC-0811-28: AI requests now carry a service deadline: a 429 refusal instead of a hang
2026-08-10 NEW-0810-16: agent model bitrix/bitrixgpt-5.6-agent
2026-08-10 NEW-0810-17: bitrix/bitrixgpt-5.5-agent is deprecated
2026-08-06 NEW-0806-1: 402 for an exhausted Cowork/Code quota carries a Retry-After header
2026-08-06 BC-0806-20: quota consumption is reported as percentages only
2026-08-04 FIX-0804-11: model-unavailable refusal is now 429 with a wait hint, not 502
2026-08-01 FIX-0801-1: AI endpoint response bodies no longer carry platform-internal fields
2026-07-27 FIX-0727-12: Connect keys: stored scopes are authoritative — vibe:ai / vibe:search are no longer added automatically
2026-07-25 NEW-0725-2: model calls with a short-lived token issued to a partner system
2026-07-21 BC-0721-10: a broken image_url candidate no longer fails the whole request
2026-07-20 NEW-0720-3: streaming chat-completions now ends a stalled upstream response with an explicit error
2026-07-16 BC-0716-1: json_object on a reasoning model recovers JSON; the 422 body is refined
2026-07-12 NEW-0712-1: The AI quota exhaustion response now points to the top-up path
2026-07-10 FIX-0710-6: non-streaming chat/completions and embeddings no longer abort at 360 seconds
2026-07-10 NEW-0710-16: AI quota pacing: pacing field in the response and 429 ai_pacing_limited error
2026-07-08 BC-0708-7: structured output: a truncated or empty result now returns 422 instead of an empty 200
2026-07-06 NEW-0706-2: New 402 error code ai_quota_exhausted on AI endpoints
2026-07-06 FIX-0706-3: Over-quota AI usage is charged at the model's base catalog price
2026-07-05 NEW-0705-5: New 402 error code ai_quota_exhausted on AI endpoints
2026-07-03 FIX-0703-11: Omitting the model field in chat again falls back to the default model