// Release stream · daily archive
Thu, Oct 1
109 releases from tracked repos.
Thu, Oct 1· 76
-
stablyai/orca v1.4.218 Patch Desktop appNew agent
Thank you so much for using Orca and for your continued support! ❤️
-
n8n-io/n8n [email protected] Sub-package +1 more
2.41.5 (2026-10-01)
-
google/adk-go v2.5.0 Major
fix(internal/utils): skip nil Parts in the Parts accessors by @dhakad22klx in https://github.com/google/adk-go/pull/1556
-
huggingface/transformers v5.18.0 Minor New model
New Model additions
-
unslothai/unsloth v0.1.902-beta Patch Desktop app
This release brings faster navigation, shareable run settings, and clearer errors to Unsloth Desktop. It also 4x speeds up Laya decisions, expands hosted Decision API support, and keeps NVFP4, INT4, and MXFP4 checkpoints
-
code-yeongyu/oh-my-openagent v5.1.8 Patch Desktop app +2 more
A packaged install runs Bun itself again when you call `bun` from an eval cell or the bash tool.** Before, it started a second agent and handed back that agent's reply as a passing result. Geeky · Heavy and `deep-high` m
-
docling-project/docling v2.132.0 First release
Get your documents ready for gen AI
-
CherryHQ/cherry-studio v2.1.4 Patch Desktop appMobile
Downloads (v2.1.4)
-
career-ops-hq/career-ops career-ops-v1.35.0 Minor +1 more
This release widens where career-ops looks: nine new job sources, from the applicant systems of US school districts to OCC Mundial in Mexico. It carries work from 86 people, 50 of them contributing for the first time.
-
mvanhorn/last30days-skill v3.26.0 Minor
harden(save): sanitize --save-suffix into a path-safe token by @nate-avra in https://github.com/mvanhorn/last30days-skill/pull/736
-
anthropics/claude-code v2.1.286 Patch New agentIDE extension
Added a count such as "2 of 5" to the permission prompt when several permission requests stack up
-
twentyhq/twenty twenty/v2.44.0 Sub-package
⚠️ Breaking changes
-
sickn33/agentic-awesome-skills v18.11.0 Minor
[18.11.0] - 2026-10-01 - "Repository Engineering, Blueprint Planning, and a Cleaner Contributor Pipeline"
-
agno-agi/agno v3.1.0 Minor
New Features
-
kortix-ai/suna v0.13.45 Patch Desktop app +1 more
Session labels, a redesigned approval flow, and long sessions that compact instead of failing.
-
Yeachan-Heo/oh-my-claudecode v5.6.0 Minor
oh-my-claudecode v5.6.0: omc ralph verify, omc ralph afk, risk-ordered stories, repo
-
CopilotKit/CopilotKit v1.76.0 Minor
feat: AG-UI 1.0 for CopilotKit (#7270) (7693a04)
-
langfuse/langfuse v4.49.0 Minor
feat(web): add org-level ingestion overview tRPC route by @hassiebp in https://github.com/langfuse/langfuse/pull/18069
-
zed-industries/zed v1.22.0 Patch New agentNew model +1 more
This week's release includes BYOK support for GPT-6.1 Sol, on-the-fly model selection for new subagents, automatic subagent compaction, a unified/split toggle for diffs opened from the command line, and dynamically fetch
-
allenai/Olmo-core v3.0.0 Major
Added `use_array_if_local` to `pack_documents_into_instances`, `segment_documents_into_instances`, `NumpyPackedFSLDataset` and `NumpyPackedFSLDatasetConfig`, forwarded to `iter_document_indices`. Set it to `False` to tak
-
modelcontextprotocol/inspector 2.5.0 First release +4 more
Visual testing tool for MCP servers
-
affaan-m/ECC v2.2.2 Patch
ECC 2.2.2 adds `pi/core`, a curated Pi-native skills+prompts-only profile built for
-
osaurus-ai/osaurus 0.25.16 Patch Desktop app
Pin MLX 0.32.3 for the crash-fix release (#2956) by @jjang-ai
-
microsoft/agent-framework dotnet-1.23.0 Minor
be01deccaf3dd7b85e03620145bc1a757c857827 .NET: Add origin pinning to Foundry toolbox MCP client (#8721)
-
anomalyco/opencode v1.18.34 Patch
Send namespaced session and parent-session identity headers with model requests.
-
Comfy-Org/ComfyUI_frontend v1.56.1 Minor New agent +3 more
fix: reinstate pointer-events on card "..." popover under a modal Reka dialog by @huntcsg in https://github.com/Comfy-Org/ComfyUI_frontend/pull/17933
-
Arize-ai/phoenix @arizeai/[email protected] Sub-package +4 more
AI Observability & Evaluation
-
jo-inc/camofox-browser v1.18.0 Minor
Stealth headless browser for AI agents — bypass Cloudflare, bot detection, and anti-scraping. Drop-in Puppeteer/Playwright replacement.
-
triton-inference-server/server v2.73.0 Minor
Copyright (c) 2025-2026, NVIDIA CORPORATION. All rights reserved.
-
Kilo-Org/kilocode v7.8.3 Patch New agent +1 more
Kilo is the all-in-one agentic engineering platform. Build, ship, and iterate faster with the most popular open source coding agent.
-
OpenWhispr/openwhispr windows-system-audio-helper-v1.2.0 Minor
Prebuilt Windows system audio helper binary for meeting transcription.
-
zhouxiaoka/autoclip v1.5.0 Minor
AutoClip|一个链接,一键出片。开源 AI 视频剪辑桌面工具,将播客、访谈、课程等长视频自动剪成短视频,生成字幕、封面和发布文案,适配抖音、小红书、TikTok、Reels 与 YouTube Shorts。Open-source AI video clipping & content repurposing.
-
siyuan-note/siyuan v3.8.7-alpha.2 Patchpre Desktop app
Improve document scrolling past fully visible mind maps
-
pytorch/pytorch v2.14.1 Patch
This release is meant to fix the following regressions and silent correctness issues:
-
pbakaus/impeccable engine-v0.1.9 Patch
Prebuilt impeccable engine binaries (engine-v0.1.9). The launcher, the npm shim and impeccable install download these on first run. Docs: https://impeccable.style
-
cline/cline desktop-v0.0.40 Patch +2 more
Custom providers added with **Add Provider** now work when you run a task. They showed up in the provider and model pickers but failed with `Unknown or disabled provider`
-
openinterpreter/openinterpreter rust-v0.0.55 Patch +1 more
Open Interpreter 0.0.55
-
AstrBotDevs/AstrBot v4.29.0-beta.1 Minorpre
AI Agent Assistant & development framework that integrates lots of IM platforms, LLMs, plugins and AI feature, and can be your openclaw alternative.
-
googleapis/python-genai v2.26.0 Minor
2.26.0 (2026-09-30)
-
huggingface/huggingface_hub v2.1.0 Minor +1 more
📊 Jobs: retries, rescheduling, and port exposure
-
anthropics/anthropic-sdk-python v1.11.0 Minor +1 more
(2026-09-30)
-
meilisearch/meilisearch v1.52.4 Patch +2 more
A lightning-fast search engine API bringing AI-powered hybrid search to your sites and applications.
-
agent-substrate/substrate v0.3.0 Minor
Update agentgateway for split actor and ateom identities by @EItanya in https://github.com/agent-substrate/substrate/pull/1923
-
multica-ai/multica v0.6.1 Patch
d61a2073bad0fbb7140fd2000f1ab7ec0a4f25 MUL-7780 feat(issues): run charts follow the pointer; a cost sparkline replaces the sidebar strip (#8912)
-
tinyhumansai/openhuman v0.64.10 Patch
The Speed, Structure & Reliability Upgrade
-
iOfficeAI/OfficeCLI v1.0.153 Patch
fix(xlsx): the sheet help schema declares the set operations the handler applies by @argszero in https://github.com/iOfficeAI/OfficeCLI/pull/440
-
Niko1221/Strata v0.1.31 Patch +1 more
Unsloth's Q4 model runs on a 64 GB PC (experimental), AMD decode up to 15% faster, models load twice as fast on Windows, and a batch of fixes.**
-
langchain-ai/deepagents deepagents==0.7.21 Patch
`FilesystemMiddleware` can now store blobs on the backend (#6550).
-
opendataloader-project/opendataloader-pdf v2.5.12 Patch
chore: begin 2.5.12-SNAPSHOT by @bundolee in https://github.com/opendataloader-project/opendataloader-pdf/pull/746
-
LibreChat-AI/LibreChat v0.8.8 Patchpre New agent
https://www.librechat.ai/changelog/v0.8.8
-
gastownhall/beads v1.3.1 Patch
Beads - A memory upgrade for your coding agent
-
deepset-ai/haystack v3.3.0 Patch +1 more
Open-source AI orchestration framework for building context-engineered, production-ready LLM applications. Design modular pipelines and agent workflows with explicit control ove…
-
earthtojake/text-to-cad v0.7.5 Patch
Other Changes
-
huggingface/peft v0.21.2 Patch
This is a PEFT release fixes an issue that prevented encoder-decoder models to work when using Transformers ≥ 5.18.0.
-
ml-explore/mlx-swift 0.32.2 Minor +1 more
Swift API for MLX
-
lexmount/moli v1.1.12 Patch
CDP, DevTools & Protocol
-
pytorch/vision v0.29.1 Patch
This is a patch release, which is compatible with PyTorch 2.14.1. There are no new features added.
-
YishenTu/claudian 2.3.11 Patch
feat(codex): mirror session archive and restore to Codex threads (#1496)
-
llmware-ai/llmware v0.4.7 Patch
llmware 0.4.7 release wheel includes prepackaged, preconfigured llama cpp binaries for multiple optimized environments 'out of the box', including: DGX Spark, AMD GPU, Windows CUDA (4090+), Linux CUDA (8.9+), along with
-
anthropics/claude-agent-sdk-typescript v0.3.286 Patch New agent
Added the initialize response field `sdk_mcp_manifests_parked` and the `system/init` capabilities `sdk_mcp_manifests` and `sdk_mcp_tools_list_changed`
-
openvinotoolkit/openvino 2026.4.1 Rolling New model
Summary of major features and improvements
-
google-gemini/gemini-cli v0.64.0-nightly.20261001.gc6bccb7ec Patchpre
fix(cli): prevent CPU hang and quote swallowing on @ within code (#29434) by @elberthc-byte in https://github.com/google-gemini/gemini-cli/pull/29557
-
mobile-next/mobile-mcp 1.0.6 Patch
Model Context Protocol Server for Mobile Automation and Scraping (iOS, Android, Emulators, Simulators and Real Devices)
-
cloudflare/computer @cloudflare/[email protected] Sub-package
Give your agent a computer
-
superradcompany/microsandbox sdk%2Fgo%2Fv0.7.6 Patch +3 more
easy, fast, programmable and local-first microVM runtime and library
-
anthropics/claude-agent-sdk-python v0.2.163 Patch
Internal/Other Changes
-
mvschwarz/openrig v0.6.3 Patch
Published:** `@openrig/[email protected]` is available on npm as `latest`.
-
Tencent/AI-Infra-Guard v4.6.4 Patch
[v4.6.4] - 2026-10-01
-
linkedin/Liger-Kernel v0.8.4 Patch
Efficient Triton Kernels for LLM Training
-
Gaurav-Gosain/tuios v0.8.4 Patch +1 more
tuios 0.8.4 is a fix release. It repairs the default mode of `tuios xpanes` and removes more invisible characters from untrusted text. It also fixes the Homebrew cask warning and a race in the herdr socket.
-
aipoch/open-science v0.34.1 Patch
AIPOCH Open-Science v0.34.1
-
OpenBMB/PilotDeck v2026.10.01 Rolling
Task-oriented AI Agent productivity platform
-
kvcache-ai/AgentENV v0.2.3 Patch
📖 Documentation for this release: https://kvcache-ai.github.io/AgentENV/v0.2.3/
-
ROCm/FastFlowLM v1.0.7 Patch New model
This release adds **`qwen3.8-mtp:27b`** — the largest dense model FastFlowLM has shipped to date, and the first to run **speculative decoding** on the NPU through its built-in MTP draft head. It also renames `hy-mt2:1.8b
-
paradedb/paradedb v0.26.0-rc.4 Patchpre +1 more
chore: Generate Dockerfiles for v0.25.0 by @paradedb-github-bot[bot] in https://github.com/paradedb/paradedb/pull/5731
-
flashinfer-ai/flashinfer v0.7.1rc1 Patchpre
feat(comm): copy-engine based kernel for the PCIe IPC all-reduce by @qsang-nv in https://github.com/flashinfer-ai/flashinfer/pull/4870
Nothing matches this filter.