# Kaeso 0.5.6

Veröffentlicht am 2. November 2026, mit 47 Änderungen.

Webseite: https://kaeso.app/de/changelog/0.5.6
In anderen Sprachen: [English](https://kaeso.app/changelog/0.5.6.md) · [简体中文](https://kaeso.app/zh-CN/changelog/0.5.6.md)

Die Release-Notizen sind auf Englisch.

## Hinzugefügt

- Cerebras is a built-in provider: add its key and call its models from any app format ([#642](https://github.com/logicplanes/kaeso/pull/642))

## Geändert

- The model API takes request bodies up to 32 MiB by default, as Anthropic does, instead of 8 MiB ([#617](https://github.com/logicplanes/kaeso/pull/617))
- Kaeso decodes large requests about a third faster by counting each body once ([#623](https://github.com/logicplanes/kaeso/pull/623))

### Gateway

- The gateway serves more requests per second under heavy load and spends less CPU on each ([#600](https://github.com/logicplanes/kaeso/pull/600))
- A response cache lookup on a large request no longer holds up other requests on the same gateway thread ([#626](https://github.com/logicplanes/kaeso/pull/626))
- A fresh install starts with KPMR's OpenRouter prices as of 2 November 2026, 22:15 UTC ([#644](https://github.com/logicplanes/kaeso/pull/644))

### KPP

- A Chat Completions tool call streamed without an id no longer slows the stream down as it grows ([#615](https://github.com/logicplanes/kaeso/pull/615))
- Kaeso translates large Responses requests, such as long Codex threads, with less work ([#627](https://github.com/logicplanes/kaeso/pull/627))
- Kaeso translates streamed replies with less work per chunk, 1.3 to 1.6 times faster for the common pairs ([#641](https://github.com/logicplanes/kaeso/pull/641))

### Self-hosting

- Updates finish 10 to 15 seconds sooner, because new containers report healthy as soon as they are ([#636](https://github.com/logicplanes/kaeso/pull/636))

## Behoben

- Streamed tool calls keep the tool name the app defined ([#598](https://github.com/logicplanes/kaeso/pull/598))
- Shortening an organization's log retention asks first, and saving one retention keeps an unsaved edit of the other ([#602](https://github.com/logicplanes/kaeso/pull/602))
- Streams that break off log the tool call Kaeso was still holding back and count its tokens in the usage estimate ([#603](https://github.com/logicplanes/kaeso/pull/603))
- Tool call arguments, citations and unknown parts that a Responses stream sends only in the finished item reach every app ([#606](https://github.com/logicplanes/kaeso/pull/606))
- A system or developer message between two tool results reaches Chat Completions models after the last result ([#608](https://github.com/logicplanes/kaeso/pull/608))
- An Anthropic stream error in another shape or in plain text keeps the provider's message ([#609](https://github.com/logicplanes/kaeso/pull/609))
- An ended session sends you to sign in from any page, the request log stops refetching a page that keeps failing, and the project overview explains what couldn't load ([#610](https://github.com/logicplanes/kaeso/pull/610))
- A Chat Completions stream's second finish chunk no longer adds an empty text block after the stop ([#612](https://github.com/logicplanes/kaeso/pull/612))
- Date pickers name the day you picked, and the request and audit logs filter by days in the install's time zone ([#613](https://github.com/logicplanes/kaeso/pull/613))
- A Combobox lists every option again after a pick, and a data table you can open rows in is a grid that doesn't open a row when you select its text ([#621](https://github.com/logicplanes/kaeso/pull/621))
- The project's forms show why a name was refused under the name, and the key and service account dialogs go back to it ([#624](https://github.com/logicplanes/kaeso/pull/624))
- Request logs say a stream ended because Kaeso restarted, not because the app left, and a broken-off upload is no longer called too large ([#625](https://github.com/logicplanes/kaeso/pull/625))
- The `postgres.pool_size` setting now says it sizes each instance's pool and that background jobs open up to 8 connections outside it ([#633](https://github.com/logicplanes/kaeso/pull/633))
- The gateway raises its open-file limit at startup, so heavy traffic no longer runs it out of sockets at about 500 streams ([#635](https://github.com/logicplanes/kaeso/pull/635))
- Automatic updates retry within the window's hour when they can't start, and no longer retry a release whose update rolled back ([#639](https://github.com/logicplanes/kaeso/pull/639))
- The updater gives each Docker call its own timeout, so a hung Docker no longer holds an update for 10 minutes past its health deadline ([#640](https://github.com/logicplanes/kaeso/pull/640))
- The password hint wraps under its field instead of running past the dialog and off a phone's screen ([#645](https://github.com/logicplanes/kaeso/pull/645))

### Gateway

- Long agent conversations no longer hit "busy translating" at half the memory budget ([#601](https://github.com/logicplanes/kaeso/pull/601))
- Limit checks cost the same however many requests are open, instead of slowing Valkey down as streams pile up ([#604](https://github.com/logicplanes/kaeso/pull/604))
- Limited streams no longer pause for Valkey once a minute, and a limited reply ends as soon as it is charged ([#607](https://github.com/logicplanes/kaeso/pull/607))
- Requests no longer wait for PostgreSQL every 30 seconds when a key's cached entries expire ([#616](https://github.com/logicplanes/kaeso/pull/616))
- Budgets no longer lose up to 10 seconds of usage when their counter is rebuilt during a usage write ([#619](https://github.com/logicplanes/kaeso/pull/619))
- A restart no longer loses the last seconds of usage and request logs when Docker stops the gateway before its requests have drained ([#631](https://github.com/logicplanes/kaeso/pull/631))
- Scheduling an organization's deletion no longer reads every user of the install, and no longer reports an error once the organization is scheduled ([#632](https://github.com/logicplanes/kaeso/pull/632))
- Request logs keep flowing after ClickHouse recovers from a stall, instead of the writer hanging until a restart ([#634](https://github.com/logicplanes/kaeso/pull/634))

### Console

- Instance users and organizations stay in the gateway's order instead of sorting only the rows loaded so far ([#597](https://github.com/logicplanes/kaeso/pull/597))
- The playground's model list follows catalogue changes, the audit log's Clear search removes the target filter, and deleting your account forgets what the browser kept ([#599](https://github.com/logicplanes/kaeso/pull/599))

## Sicherheit

- Changing your password can also disconnect your agents, for when someone else may have connected one ([#605](https://github.com/logicplanes/kaeso/pull/605))
- A disabled key can no longer be rotated into a working one; enable it first ([#618](https://github.com/logicplanes/kaeso/pull/618))
- Ending your own sessions is now recorded in the instance's audit log ([#620](https://github.com/logicplanes/kaeso/pull/620))
- An agent no longer sees organizations and projects it was shut out of in your memberships or a member's rights ([#622](https://github.com/logicplanes/kaeso/pull/622))
- An organization past its deletion date can no longer be restored, and a restore under way stops its final deletion instead of losing its logs ([#628](https://github.com/logicplanes/kaeso/pull/628))
- Changing an organization's log retention or content logging now needs a recent sign-in, and the right is marked sensitive ([#629](https://github.com/logicplanes/kaeso/pull/629))
- Agent sign-ins are limited per address and capped, unknown agent tokens count toward key throttling, and sign-in errors go back only to approved agents ([#638](https://github.com/logicplanes/kaeso/pull/638))

### Gateway

- An ownership transfer racing with a demotion, removal or leave no longer leaves an organization without an owner ([#611](https://github.com/logicplanes/kaeso/pull/611))
- A key created while its member is removed, its account deleted or its project or service account archived is refused or archived with the rest ([#614](https://github.com/logicplanes/kaeso/pull/614))
- A burst of sign-in attempts can no longer run the gateway out of memory ([#630](https://github.com/logicplanes/kaeso/pull/630))

[Das Release auf GitHub](https://github.com/logicplanes/kaeso/releases/tag/v0.5.6)