|
`gpt-5.4` system prompt contains invalid cutoff of 2024-06
|
|
7
|
691
|
September 8, 2026
|
|
Code Interpreter fails to surface container-mounted skills to the model, unlike the Shell tool
|
|
1
|
164
|
September 8, 2026
|
|
Realtime API interruptions are far too sensitive even at a high VAD threshold value
|
|
3
|
1505
|
September 8, 2026
|
|
Hosted web_search tool injects a UTC "current date" into the model context, overriding developer-provided local date (off-by-one for non-UTC users)
|
|
3
|
137
|
September 8, 2026
|
|
Ads Insights API time_ranges validation not in account timezone
|
|
3
|
92
|
September 8, 2026
|
|
GPT 5.6 Don't want to Work?
|
|
0
|
53
|
September 8, 2026
|
|
Structured Outputs - un-minified
|
|
2
|
136
|
September 7, 2026
|
|
Gpt-5.6 chat completion reasoning effort [bug / behavior change]
|
|
3
|
2928
|
September 7, 2026
|
|
Gpt-realtime-translate bugs & feedback
|
|
1
|
138
|
September 7, 2026
|
|
Logs not updating since Jun 23
|
|
1
|
74
|
September 7, 2026
|
|
Realtime API Webhook endpoints -> Send test event - only works sometimes
|
|
1
|
66
|
September 7, 2026
|
|
Wfr_ errors in Responses API background mode + correlation with code_interpreter-heavy workflows?
|
|
1
|
96
|
September 7, 2026
|
|
Gpt-realtime 1.5: spoken audio drops digit in long repeated-digit numeric strings (transcript correct, audio wrong)
|
|
1
|
117
|
September 7, 2026
|
|
Model returning malformed characters in JSON response using API
|
|
9
|
1022
|
September 7, 2026
|
|
Background-mode Responses stuck in "queued" indefinitely (never scheduled), reproducible with minimal requests
|
|
28
|
691
|
September 7, 2026
|
|
Realtime API SIP accept endpoint takes ~5.3s to return 404 call_id_not_found
|
|
2
|
93
|
September 6, 2026
|
|
Batch API Outage - /v1/responses cannot access valid purpose=batch input_file_id
|
|
317
|
7097
|
September 4, 2026
|
|
/v1/responses with background: true does not respect the json_schema (with strict: true), works when background:false
|
|
0
|
35
|
September 4, 2026
|
|
Gpt-5.6: usage.output_tokens is ~9x actual generation (re-summed once per reasoning item), exceeds max_output_tokens, and is what gets billed
|
|
25
|
2819
|
September 3, 2026
|
|
Realtime /v1/realtime/calls suddenly failing with 401 ip_not_authorized despite static Cloud Run IP being allowlisted
|
|
1
|
116
|
September 2, 2026
|
|
Responses API + Structured Outputs (gpt-5.6-luna): garbage tokens (foreign scripts / leaked reasoning) inside string values right before the closing quote — identical request via Chat Completions is clean
|
|
4
|
558
|
September 2, 2026
|
|
5.6 Luna schema bug when called on OpenRouter
|
|
0
|
62
|
August 31, 2026
|
|
Realtime API over SIP: calls always terminated at ~30.16s, sideband WebSocket never receives a single event
|
|
0
|
76
|
August 30, 2026
|
|
PDF file uploaded to vector store cannot be used by the model
|
|
7
|
121
|
August 30, 2026
|
|
Bug Report: Persistent "Something went wrong. Please try again." streaming failure
|
|
2
|
78
|
August 28, 2026
|
|
Batch images edits failing with 401 Unable to authorize file access
|
|
22
|
890
|
August 28, 2026
|
|
424 Failed Dependency error when connecting to Stripe remote MCP connector
|
|
1
|
90
|
August 28, 2026
|
|
OpenAI Batch API failing since 19th August
|
|
1
|
868
|
August 28, 2026
|
|
Organization rate limit bug: completion internal model names with "not available" showing
|
|
4
|
141
|
August 26, 2026
|
|
Max_tool_calls exceeded: the counter appears to reset after compaction
|
|
3
|
173
|
August 26, 2026
|