Skip to content

deps: bump everos from 1.2.3 to 1.4.1 - #218

Merged
RonsenbergVI merged 1 commit into
mainfrom
dependabot/uv/everos-1.4.1
Oct 10, 2026
Merged

deps: bump everos from 1.2.3 to 1.4.1#218
RonsenbergVI merged 1 commit into
mainfrom
dependabot/uv/everos-1.4.1

Conversation

@dependabot

@dependabot dependabot Bot commented on behalf of github Oct 9, 2026

Copy link
Copy Markdown
Contributor

Bumps everos from 1.2.3 to 1.4.1.

Release notes

Sourced from everos's releases.

EverOS 1.4.1

A fresh install works again. The openai SDK released its 3.x line on 2026-09-24; an unconstrained install picked it up together with an httpx pre-release, and every LLM and embedding call failed before reaching the network. This release pins the SDK to the 2.x line the project is tested against. Nothing else changed since 1.4.0.

Fixed

  • openai is pinned below 3. Fresh installs from PyPI resolved openai 3.19.2, whose client raises AttributeError: module 'httpx' has no attribute 'Timeout' on every request, so memorize and hybrid / vector search returned 500. Existing environments built from uv.lock were never affected.

Upgrade

pip install --upgrade everos   # or: uv sync
  • pip install --upgrade everos brings openai back to 2.x. If a fresh install of 1.4.0 (or any earlier version) today shows the httpx error above, upgrade to 1.4.1 or run pip install "openai<3".

Full changelog: v1.4.0...v1.4.1

EverOS 1.4.0

EverOS runs natively on Windows, and dense search stops scanning the whole table. pip install everos on a stock Windows 11 machine now works with no manual steps beyond an LLM key; the unit suite runs on Windows in CI. Every vector column gets an IVF_FLAT index once it holds enough rows, built and maintained by the cascade worker, so hybrid search no longer reads the entire column per query. A ten-hour Windows soak drove the rest of this release: a second process can no longer write the index behind a running server, watcher events for one file commit in the order they arrived, and a transient Lance spill failure is retried instead of filing the markdown file as unrecoverable.

Added

  • Native Windows support. The package ships the MSVC runtime (msvc-runtime, Windows only) and registers its directory before any native extension loads, so a clean machine needs no Visual C++ Redistributable. Text fixtures are read as UTF-8 regardless of the system locale, and a unit tests (Windows) CI job covers the platform. https://github.com/EverMind-AI/EverOS/blob/HEAD/docs/windows.md leads with the native install and keeps the WSL2 walkthrough.
  • Vector (ANN) indexes. Once a vector column holds [lancedb] vector_index_min_rows (default 2000) non-null vectors, the cascade worker builds an IVF_FLAT cosine index on it — on its first rebuild

... (truncated)

Changelog

Sourced from everos's changelog.

[1.4.1] - 2026-09-24

A fresh install works again. The openai SDK released its 3.x line on 2026-09-24; an unconstrained install picked it up together with an httpx pre-release, and every LLM and embedding call failed before reaching the network. This release pins the SDK to the 2.x line the project is tested against. Nothing else changed since 1.4.0.

Fixed

  • openai is pinned below 3. Fresh installs from PyPI resolved openai 3.19.2, whose client raises AttributeError: module 'httpx' has no attribute 'Timeout' on every request, so memorize and hybrid / vector search returned 500. Existing environments built from uv.lock were never affected.

Upgrade

  • pip install --upgrade everos brings openai back to 2.x. If a fresh install of 1.4.0 (or any earlier version) today shows the httpx error above, upgrade to 1.4.1 or run pip install "openai<3".

[1.4.0] - 2026-09-24

EverOS runs natively on Windows, and dense search stops scanning the whole table. pip install everos on a stock Windows 11 machine now works with no manual steps beyond an LLM key; the unit suite runs on Windows in CI. Every vector column gets an IVF_FLAT index once it holds enough rows, built and maintained by the cascade worker, so hybrid search no longer reads the entire column per query. A ten-hour Windows soak drove the rest of this release: a second process can no longer write the index behind a running server, watcher events for one file commit in the order they arrived, and a transient Lance spill failure is retried instead of filing the markdown file as unrecoverable.

Added

  • Native Windows support. The package ships the MSVC runtime (msvc-runtime, Windows only) and registers its directory before any native extension loads, so a clean machine needs no Visual C++ Redistributable. Text fixtures are read as UTF-8 regardless of the system locale, and a unit tests (Windows) CI job covers the platform. https://github.com/EverMind-AI/EverOS/blob/main/docs/windows.md leads with the native install and keeps the WSL2 walkthrough.
  • Vector (ANN) indexes. Once a vector column holds [lancedb] vector_index_min_rows (default 2000) non-null vectors, the cascade worker builds an IVF_FLAT cosine index on it — on its first rebuild sweep after server start and on the heavy maintenance beat — and retrains it once light-beat delta indices pile past 16. Partition size (4096 rows) and nprobes (32) are pinned together, so the search is exact up to ~130k rows. Measured on 27k rows of 1024-dim vectors: 245 ms → 84 ms per query with

... (truncated)

Commits
  • 462ebf9 fix(deps): pin openai below 3 so a fresh install can call the LLM
  • 1fd0db1 chore(release): v1.4.0
  • 71517c8 chore(release): v1.4.0rc2
  • 8bb5c32 fix(lancedb): translate the spill failure on the write path too
  • f00161b chore(release): v1.4.0rc1
  • 4e905dc feat(lancedb): build an IVF_FLAT index on vector columns
  • ca6ea7d feat(windows): unblock the import path and cover it in CI
  • 4fd3472 fix(lancedb): treat a lance spill failure as retryable
  • 401fbf8 fix(knowledge): delete the document directory even before it is indexed
  • b736df6 build(deps): bump pyarrow to 25.0.1 for timestamp materialisation
  • Additional commits viewable in compare view

Dependabot compatibility score

Dependabot will resolve any conflicts with this PR as long as you don't alter it yourself. You can also trigger a rebase manually by commenting @dependabot rebase.

Dependabot commands and options

You can trigger Dependabot actions by commenting on this PR:

  • @dependabot rebase will rebase this PR
  • @dependabot recreate will recreate this PR, overwriting any edits that have been made to it
  • @dependabot show <dependency name> ignore conditions will show all of the ignore conditions of the specified dependency
  • @dependabot ignore this major version will close this PR and stop Dependabot creating any more for this major version (unless you reopen the PR or upgrade to it yourself)
  • @dependabot ignore this minor version will close this PR and stop Dependabot creating any more for this minor version (unless you reopen the PR or upgrade to it yourself)
  • @dependabot ignore this dependency will close this PR and stop Dependabot creating any more for this dependency (unless you reopen the PR or upgrade to it yourself)

@dependabot
deps: bump everos from 1.2.3 to 1.4.1
73ab548 
Bumps [everos](https://github.com/EverMind-AI/EverOS) from 1.2.3 to 1.4.1.
- [Release notes](https://github.com/EverMind-AI/EverOS/releases)
- [Changelog](https://github.com/EverMind-AI/EverOS/blob/main/CHANGELOG.md)
- [Commits](EverMind-AI/EverOS@v1.2.3...v1.4.1)

---
updated-dependencies:
- dependency-name: everos
  dependency-version: 1.4.1
  dependency-type: direct:production
  update-type: version-update:semver-minor
...

Signed-off-by: dependabot[bot] <support@github.com>
@dependabot dependabot Bot added dependencies Pull requests that update a dependency file python:uv Pull requests that update python:uv code labels Oct 9, 2026
@RonsenbergVI
RonsenbergVI merged commit ff7c244 into main Oct 10, 2026
5 checks passed
@dependabot
dependabot Bot deleted the dependabot/uv/everos-1.4.1 branch October 10, 2026 18:17
@RonsenbergVI RonsenbergVI mentioned this pull request Oct 9, 2026
Merged
RonsenbergVI added a commit that referenced this pull request Oct 20, 2026
@github-actions @RonsenbergVI
chore(main): release 0.1.0 (#32)
18d41e8 
🤖 I have created a release *beep* *boop*
---


##
[0.1.0](v0.0.1...v0.1.0)
(2026-10-20)


### ⚠ BREAKING CHANGES

* agent-only runs scored by judge accuracy, on Anthropic
([#242](#242))
* remove the retired letta integration
([#238](#238))

### Features

* add langmem benchmark
([#166](#166))
([f385975](f385975))
* add LongMemEval xs and xs-shared variants, cut from s at load time
([#241](#241))
([bdb9543](bdb9543))
* add memos benchmark
([#165](#165))
([b5b4701](b5b4701))
* agent-only runs scored by judge accuracy, on Anthropic
([#242](#242))
([ab65ceb](ab65ceb))
* **agentmemory:** add agentmemory integration
([#118](#118))
([b3e16bf](b3e16bf))
* **cognee:** add cognee integration
([#113](#113))
([b258d06](b258d06))
* distinguish partial token accounting from full and from none
([#124](#124))
([3129e34](3129e34))
* **everos:** add everos integration
([#121](#121))
([61832f8](61832f8))
* **fraise:** expose one raw FQL query tool, with the language spec
([#254](#254))
([c485835](c485835))
* **hindsight:** add hindsight integration
([#120](#120))
([78337e8](78337e8))
* **letta:** add agent-driven ingestion that exercises the LLM
([#122](#122))
([6ad2d55](6ad2d55))
* LongMemEval xs at LoCoMo's ingest time, with per-system workflows
([#249](#249))
([93e417e](93e417e))
* mark systems whose token spend cannot be observed
([#123](#123))
([4c157e1](4c157e1))
* **reporting:** attribute each miss to the memory or the agent, and
score the memory's lift
([#265](#265))
([c2fc6b1](c2fc6b1))
* **reporting:** rank the summary by retrieval recall
([#262](#262))
([da63445](da63445))


### Bug fixes

* count tokens spent off the worker thread, and async chat tokens
([#119](#119))
([2a324c8](2a324c8))
* **everos:** restore the everalgo-boundary &lt;0.3 cap
([#253](#253))
([53ca990](53ca990))
* **fraise:** phrase-only recall, SDK 0.1.0b3, and no session header
([#102](#102))
([4833c21](4833c21))
* **graphiti:** document the raised llm token ceiling
([#90](#90))
([3302d5f](3302d5f))
* **graphiti:** drop truststore from the TLS path
([#100](#100))
([c177b95](c177b95))
* **graphiti:** stop closing async clients in teardown
([#94](#94))
([a8d5fa4](a8d5fa4))
* increase line styling count
([#181](#181))
([00829ed](00829ed))
* **letta:** stop injecting a session header into stored passages
([#104](#104))
([299374b](299374b))
* match tagged toml values in release-please lockfile filters
([#92](#92))
([02bb9ef](02bb9ef))
* **mem0:** serialize client setup to avoid the qdrant collection race
([#103](#103))
([437a48f](437a48f))
* **reporting:** sweep recall, not accuracy, per question category
([#263](#263))
([6c7ec0b](6c7ec0b))


### Dependencies

* bump actions/download-artifact from 4 to 8
([#5](#5))
([fcc4acb](fcc4acb))
* bump actions/upload-artifact from 4 to 7
([#6](#6))
([a88e482](a88e482))
* bump docker/login-action from 3 to 4
([#8](#8))
([72458c3](72458c3))
* bump everalgo-boundary from 0.2.1 to 0.3.0
([#144](#144))
([8bebd3d](8bebd3d))
* bump everalgo-boundary from 0.2.1 to 0.3.0
([#156](#156))
([de299f1](de299f1))
* bump everalgo-boundary from 0.2.1 to 0.3.0
([#168](#168))
([8cdaf02](8cdaf02))
* bump everalgo-boundary from 0.2.1 to 0.3.0
([#177](#177))
([8a0ec1d](8a0ec1d))
* bump everalgo-boundary from 0.2.1 to 0.3.0
([#189](#189))
([aad9e48](aad9e48))
* bump everalgo-boundary from 0.2.1 to 0.3.0
([#196](#196))
([40458a3](40458a3))
* bump everalgo-boundary from 0.2.1 to 0.3.0
([#206](#206))
([7f035f2](7f035f2))
* bump everalgo-boundary from 0.2.1 to 0.3.0
([#215](#215))
([e7eea35](e7eea35))
* bump everalgo-boundary from 0.2.1 to 0.3.0
([#227](#227))
([dca612b](dca612b))
* bump everos from 1.2.3 to 1.3.1
([#152](#152))
([2ec392e](2ec392e))
* bump everos from 1.2.3 to 1.3.1
([#162](#162))
([7aab956](7aab956))
* bump everos from 1.2.3 to 1.3.1
([#169](#169))
([c5d6f61](c5d6f61))
* bump everos from 1.2.3 to 1.3.1
([#188](#188))
([f3de5a2](f3de5a2))
* bump everos from 1.2.3 to 1.3.1
([#193](#193))
([5192562](5192562))
* bump everos from 1.2.3 to 1.4.1
([#197](#197))
([52e94de](52e94de))
* bump everos from 1.2.3 to 1.4.1
([#212](#212))
([1868d25](1868d25))
* bump everos from 1.2.3 to 1.4.1
([#218](#218))
([ff7c244](ff7c244))
* bump everos from 1.2.3 to 1.4.2
([#259](#259))
([16fbc26](16fbc26))
* bump fastembed from 0.8.0 to 0.8.1
([#200](#200))
([9dc9702](9dc9702))
* bump fastembed from 0.8.1 to 0.9.0
([#228](#228))
([28e0466](28e0466))
* bump fraise-sdk from 0.1.0b4 to 0.1.0rc1
([#151](#151))
([bd0777d](bd0777d))
* bump graphiti-core from 0.29.3 to 0.30.1
([#148](#148))
([aef6f0b](aef6f0b))
* bump graphiti-core from 0.30.1 to 0.30.2
([#153](#153))
([9642e65](9642e65))
* bump graphiti-core from 0.30.2 to 0.31.1
([#277](#277))
([872fe3a](872fe3a))
* bump hindsight-client from 0.10.0 to 0.10.1
([#194](#194))
([47f8390](47f8390))
* bump hindsight-client from 0.10.1 to 0.10.2
([#208](#208))
([cbd97aa](cbd97aa))
* bump hindsight-client from 0.10.2 to 0.10.3
([#225](#225))
([750b517](750b517))
* bump huggingface-hub from 1.27.0 to 1.28.0
([#110](#110))
([229af82](229af82))
* bump huggingface-hub from 1.30.0 to 1.31.0
([#154](#154))
([f366a6d](f366a6d))
* bump langgraph-checkpoint-postgres from 3.1.2 to 3.1.3
([#230](#230))
([752ea49](752ea49))
* bump langgraph-checkpoint-postgres from 3.1.3 to 3.2.0
([#256](#256))
([dc1b34f](dc1b34f))
* bump matplotlib from 3.11.1 to 3.11.2
([#161](#161))
([16251aa](16251aa))
* bump openai from 2.54.0 to 3.1.0
([#7](#7))
([38e9d7e](38e9d7e))
* bump openai from 3.1.0 to 3.2.0
([#85](#85))
([de3ef85](de3ef85))
* bump pydantic from 2.13.4 to 2.13.5
([#146](#146))
([715928f](715928f))
* bump pydantic from 2.13.5 to 2.14.1
([#229](#229))
([e135470](e135470))
* bump pydantic-ai from 2.31.0 to 2.31.1
([#37](#37))
([980ce8a](980ce8a))
* bump pydantic-ai from 2.31.1 to 2.33.0
([#112](#112))
([802ad6a](802ad6a))
* bump pydantic-ai from 2.44.0 to 2.45.0
([#187](#187))
([c4cb7b2](c4cb7b2))
* bump pydantic-ai from 2.45.0 to 2.46.0
([#191](#191))
([1a81c05](1a81c05))
* bump pydantic-ai from 2.46.0 to 2.47.0
([#195](#195))
([37c471c](37c471c))
* bump pydantic-ai from 2.47.0 to 2.50.0
([#199](#199))
([de91847](de91847))
* bump pydantic-ai from 2.50.0 to 2.51.0
([#201](#201))
([9354f79](9354f79))
* bump pydantic-ai from 2.51.0 to 2.53.0
([#213](#213))
([dafb162](dafb162))
* bump pydantic-ai from 2.53.0 to 2.54.0
([#214](#214))
([78d6b4a](78d6b4a))
* bump pydantic-ai from 2.54.0 to 2.55.0
([#224](#224))
([75653af](75653af))
* bump pydantic-ai from 2.55.0 to 2.56.0
([#260](#260))
([2debd3f](2debd3f))
* bump pydantic-ai from 2.56.0 to 2.57.0
([#276](#276))
([6a01934](6a01934))
* bump pydantic-settings from 2.15.0 to 2.16.0
([#258](#258))
([20d5b6b](20d5b6b))
* bump stefanzweifel/git-auto-commit-action from 5 to 7
([e843a8d](e843a8d))
* bump stefanzweifel/git-auto-commit-action from 5 to 7
([#11](#11))
([d189daa](d189daa))
* bump tiktoken from 0.13.0 to 0.14.0
([#38](#38))
([1aaced6](1aaced6))
* bump tqdm from 4.70.0 to 4.70.1
([#155](#155))
([1fae94a](1fae94a))
* **hindsight:** bump server and client together to 1.0.0
([#257](#257))
([04a5b54](04a5b54))
* update cognee[neo4j,postgres-binary] requirement from &lt;1.6,&gt;=1.5
to &gt;=1.5,&lt;1.7
([#190](#190))
([a48c9d9](a48c9d9))


### Maintenance

* add citation
([#101](#101))
([e2634a7](e2634a7))
* add dataset tiering
([#115](#115))
([1cf79ec](1cf79ec))
* add missing benchmark status badges
([#136](#136))
([b6b4d7e](b6b4d7e))
* add missing openai api key
([#137](#137))
([3254a0e](3254a0e))
* add question category plots
([#186](#186))
([e8ed472](e8ed472))
* add run freshness in plots
([#88](#88))
([322c1a9](322c1a9))
* add sponsorship documents
([#114](#114))
([1224626](1224626))
* CI run improvements
([#96](#96))
([d9accd3](d9accd3))
* drop component from release tags
([#17](#17))
([2f50cea](2f50cea))
* fix failed bench run
([#157](#157))
([0351636](0351636))
* **fraise:** move to server v0.2.2 and SDK 0.1.0rc2
([#203](#203))
([d05a949](d05a949))
* **fraise:** move to server v0.3.0 and SDK 0.1.1
([#210](#210))
([ca49452](ca49452))
* **fraise:** move to server v0.3.2
([#219](#219))
([1709579](1709579))
* **fraise:** move to server v0.3.3 and SDK 0.2.0
([#221](#221))
([22e55dd](22e55dd))
* **fraise:** move to v0.1.0-rc.1 and SDK 0.1.0b4
([#127](#127))
([77ad453](77ad453))
* **fraise:** update fraise docker image version
([#106](#106))
([6b9b41d](6b9b41d))
* **main:** release agentmemory 0.1.0
([#133](#133))
([036610f](036610f))
* **main:** release agentmemory 0.2.0
([#139](#139))
([53c5be4](53c5be4))
* **main:** release agentmemory 0.3.0
([#160](#160))
([ba63ef5](ba63ef5))
* **main:** release cognee 0.1.0
([#132](#132))
([fb9cab0](fb9cab0))
* **main:** release cognee 0.2.0
([#192](#192))
([e4772ab](e4772ab))
* **main:** release everos 0.1.0
([#135](#135))
([9e60001](9e60001))
* **main:** release everos 0.2.0
([#145](#145))
([adb83f9](adb83f9))
* **main:** release everos 0.3.0
([#247](#247))
([67aea45](67aea45))
* **main:** release everos 0.4.0
([#255](#255))
([3156dce](3156dce))
* **main:** release fraise 0.1.0
([#36](#36))
([7b845e7](7b845e7))
* **main:** release fraise 0.2.0
([#128](#128))
([1996c55](1996c55))
* **main:** release fraise 0.3.0
([#143](#143))
([67a0972](67a0972))
* **main:** release fraise 0.4.0
([#204](#204))
([5dc5d21](5dc5d21))
* **main:** release fraise 0.5.0
([#211](#211))
([3adaae5](3adaae5))
* **main:** release fraise 0.6.0
([#220](#220))
([4611bbb](4611bbb))
* **main:** release fraise 0.7.0
([#222](#222))
([17eaaeb](17eaaeb))
* **main:** release fraise 0.8.0
([#240](#240))
([f12fe75](f12fe75))
* **main:** release graphiti 0.1.0
([#34](#34))
([bc22102](bc22102))
* **main:** release graphiti 0.2.0
([#93](#93))
([d4ce825](d4ce825))
* **main:** release graphiti 0.3.0
([#95](#95))
([cbb3f28](cbb3f28))
* **main:** release graphiti 0.4.0
([#149](#149))
([6c269ad](6c269ad))
* **main:** release graphiti 0.5.0
([#245](#245))
([972d700](972d700))
* **main:** release hindsight 0.1.0
([#134](#134))
([32b12cc](32b12cc))
* **main:** release hindsight 0.2.0
([#138](#138))
([990a991](990a991))
* **main:** release hindsight 0.3.0
([#158](#158))
([40dc8dc](40dc8dc))
* **main:** release hindsight 0.4.0
([#239](#239))
([204af1c](204af1c))
* **main:** release hindsight 0.5.0
([#275](#275))
([ee7e700](ee7e700))
* **main:** release langmem 0.1.0
([#180](#180))
([2a15b28](2a15b28))
* **main:** release langmem 0.2.0
([#244](#244))
([421a2d7](421a2d7))
* **main:** release letta 0.1.0
([#33](#33))
([bfbb05d](bfbb05d))
* **main:** release letta 0.2.0
([#105](#105))
([28aace6](28aace6))
* **main:** release mem0 0.1.0
([#35](#35))
([f068ae4](f068ae4))
* **main:** release mem0 0.2.0
([#109](#109))
([fe5da80](fe5da80))
* **main:** release mem0 0.3.0
([#243](#243))
([f3c5776](f3c5776))
* **main:** release memos 0.1.0
([#176](#176))
([30dce41](30dce41))
* **main:** release memos 0.2.0
([#246](#246))
([f7aaad4](f7aaad4))
* pause LongMemEval across the benchmarked systems
([#126](#126))
([bcfd888](bcfd888))
* polish readme
([#86](#86))
([65a0715](65a0715))
* preparing first full run
([#19](#19))
([cd91b4b](cd91b4b))
* remove agentmemory harness
([#164](#164))
([2cea089](2cea089))
* remove candidate section
([#167](#167))
([a0dc301](a0dc301))
* remove individual memory dependabot
([#39](#39))
([a28f25b](a28f25b))
* remove the retired letta integration
([#238](#238))
([14467b0](14467b0))
* repo hygiene - first full run prep
([#31](#31))
([2a1bc04](2a1bc04))
* serialize release-please runs so a release merge isn't re-proposed
([#261](#261))
([d870852](d870852))
* update agentmemory run
([#159](#159))
([8562edb](8562edb))
* update clean target
([#207](#207))
([9aa2360](9aa2360))
* update run artefact saving
([#98](#98))
([87064e5](87064e5))
* upgrade fraise version and use one fraise graph per conversation
([#142](#142))
([8865abb](8865abb))

---
This PR was generated with [Release
Please](https://github.com/googleapis/release-please). See
[documentation](https://github.com/googleapis/release-please#release-please).

Co-authored-by: René-Jean Corneille <14303619+RonsenbergVI@users.noreply.github.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

No reviews

Assignees

No one assigned

Labels

dependencies Pull requests that update a dependency file python:uv Pull requests that update python:uv code

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

1 participant

@RonsenbergVI