open-webui/backend/open_webui/utils
Classic298 e88e1f8119
fix: prompt cache misses after a model calls tools one after another (#31593)
When a model called one tool, got its result and then called a second tool with no text in between, the next request merged both calls into an earlier assistant message that the provider had already seen. That changed the conversation's beginning, so the provider's prompt cache stopped matching from there for the rest of the chat. Each tool call and its result are now sent as their own messages, so the start of the conversation stays identical from one request to the next.

Fixes #31588
2026-10-01 07:34:59 +04:00
..
access_control chore: format 2026-09-21 11:35:30 -04:00
images fix: show ComfyUI images saved by the Save Image (Advanced) node (#30420) 2026-09-23 23:54:58 -04:00
mcp fix: MCP tool calls ignore AIOHTTP_CLIENT_TIMEOUT_TOOL_SERVER (#31641) 2026-09-30 19:00:20 +04:00
telemetry fix: every log line is exported twice to the OpenTelemetry collector when traces and logs are both on (#31528) 2026-09-30 19:36:43 +04:00
actions.py refac 2026-10-01 03:31:20 +04:00
anthropic.py fix: Anthropic API streams report a failed response as a finished one (#31405) 2026-09-27 23:20:50 +04:00
asgi_middleware.py refac 2026-08-24 18:29:36 -04:00
ask_user.py fix: a rejected ask_user call ending the turn with no reply (#29252) 2026-08-31 00:17:21 -04:00
audit.py fix: audit log shows passwords that contain a double quote (#31659) 2026-10-01 07:21:58 +04:00
auth.py refac 2026-09-19 17:54:09 -04:00
automations.py fix: check the Channels permission when a channel automation runs (#31577) 2026-09-30 19:02:37 +04:00
calendar.py fix: Scheduled Tasks calendar shows extra runs for automations with a run count (#31604) 2026-09-30 19:43:41 +04:00
channels.py refac 2026-03-17 17:58:01 -05:00
chat.py fix: keep the model system prompt on Ollama tool-call follow-ups (#30375) 2026-09-22 16:47:20 -04:00
chat_fork.py refac 2026-07-23 02:54:56 -04:00
chat_id.py refac 2026-07-26 21:12:14 -04:00
chat_variables.py fix: non-English text is missed by searches and counted six times against size limits (#31615) 2026-09-30 19:19:29 +04:00
code_interpreter.py refac 2026-07-31 17:41:14 -04:00
context_compaction.py refac 2026-08-23 13:36:53 -04:00
embeddings.py refac: modernize type annotations (PEP 604 / PEP 585) 2026-05-12 17:10:15 +09:00
files.py refac(images): request an unencoded body when fetching remote chat images (#29623) 2026-09-21 08:53:06 -04:00
filter.py refac 2026-10-01 03:31:20 +04:00
groups.py refac: modernize type annotations (PEP 604 / PEP 585) 2026-05-12 17:10:15 +09:00
headers.py refac 2026-09-27 23:29:38 +04:00
json_codec.py perf: write task payloads to Redis as bytes (#28833) 2026-08-20 12:58:52 -07:00
json_response.py perf: optional orjson JSON codec behind ENABLE_ORJSON (#27583) 2026-07-27 03:45:37 -04:00
logger.py perf: stop formatting every exported log record twice under OTEL log export (#27840) 2026-08-10 23:13:52 -06:00
memory.py fix: stop the background memory review when memory is switched off (#30309) 2026-09-21 11:22:21 -04:00
middleware.py fix: sub-agents in folder chats are not limited to the folder's knowledge (#31574) 2026-10-01 07:34:37 +04:00
misc.py fix: prompt cache misses after a model calls tools one after another (#31593) 2026-10-01 07:34:59 +04:00
model_ids.py refac 2026-07-26 23:09:22 -04:00
models.py refac 2026-10-01 03:31:20 +04:00
notifications.py fix: chat links in finished and failed webhook notifications open a 404 (#31572) 2026-09-30 19:05:45 +04:00
oauth.py refac 2026-10-01 03:31:20 +04:00
payload.py fix: max_tokens sent through the API is ignored for Ollama models (#31437) 2026-09-26 07:18:43 +04:00
plugin.py refac 2026-10-01 03:31:20 +04:00
rate_limit.py fix: stop the sign-in rate limiter blocking the loop and leaking memory (#29977) 2026-09-13 20:28:41 -05:00
recurrence.py fix: Scheduled Tasks calendar shows extra runs for automations with a run count (#31604) 2026-09-30 19:43:41 +04:00
redis.py fix: refresh an expiring OAuth token once across workers and replicas (#30450) 2026-09-23 23:33:09 -04:00
response.py refac 2026-08-17 00:57:57 -07:00
sanitize.py refac 2026-03-17 17:58:01 -05:00
security_headers.py refac 2026-08-24 18:29:36 -04:00
session_pool.py fix: long streamed lines no longer abort the response (#28114) 2026-08-25 12:16:37 -04:00
skills.py chore: format 2026-09-21 11:35:30 -04:00
subagents.py fix: sub-agents in folder chats are not limited to the folder's knowledge (#31574) 2026-10-01 07:34:37 +04:00
task.py refac 2026-06-19 00:16:06 +02:00
terminals.py refac 2026-10-01 03:31:20 +04:00
timers.py fix: background sub-agent and timer replies only show in the open chat after a page reload (#31576) 2026-09-30 19:03:01 +04:00
tool_approval.py fix: system prompt and compacted context are lost after approving a tool call (#31501) 2026-09-28 07:59:43 +04:00
tools.py refac 2026-10-01 03:31:20 +04:00
validate.py refac 2026-09-15 22:47:03 -04:00
valves.py refac 2026-07-31 17:41:14 -04:00
webhook.py perf: build debug log messages lazily so disabled debug logs cost nothing (#27834) 2026-07-31 19:09:01 -05:00