Stable releases from the AI builder tools PickOneAI ranks.https://pickoneai.com/en-ushttps://github.com/n8n-io/n8n/releases/tag/n8n%402.45.1https://github.com/n8n-io/n8n/releases/tag/n8n%402.45.12.45.1 (2026-10-21) Bug Fixes core: Claim an idle worker's next queue job and its lock in one step ( #41558 ) ( 5f0b562 ) editor: Center badge text when icons are absent ( #41581 ) ( 648fa2a ) Microsoft Teams Trigger Node: Renew the Graph subscription via lifecycle notifications Wed, 21 Oct 2026 13:00:42 GMTn8nhttps://github.com/ggml-org/llama.cpp/releases/tag/b11796https://github.com/ggml-org/llama.cpp/releases/tag/b11796mtmd: add state object for mtmd_gen ( #30902 ) Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/attestations/58202944 macOS/iOS: macOS Apple Silicon (arm64) macOS Apple Silicon (arm64, KleidiAI enabled) DISABLED macOS Intel (x64) iOS XCFramework LinuWed, 21 Oct 2026 12:39:27 GMTllama-cpphttps://github.com/ggml-org/llama.cpp/releases/tag/b11795https://github.com/ggml-org/llama.cpp/releases/tag/b11795server : forward llama.cpp sampling params on /v1/messages ( #30856 ) server : forward llama.cpp sampling params on /v1/messages server : pass the request through on /v1/messages instead of copying listed keys NIT: Change comment from multiline to single line server : accept llamWed, 21 Oct 2026 12:20:53 GMTllama-cpphttps://vercel.com/changelog/skipped-monorepo-deployments-no-longer-count-against-limitshttps://vercel.com/changelog/skipped-monorepo-deployments-no-longer-count-against-limitsDeployments that Vercel automatically skips in a monorepo no longer count against deployment rate limits.Wed, 21 Oct 2026 12:14:47 GMTvercelhttps://github.com/ggml-org/llama.cpp/releases/tag/b11793https://github.com/ggml-org/llama.cpp/releases/tag/b11793llama : optimize vocabulary loading ( #30679 ) llama : optimize vocabulary loading Update src/unicode.h Co-authored-by: Georgi Gerganov ggerganov@gmail.com CR : added reserve in gguf_kv_to_str CR : trailing space fix Co-authored-by: Georgi Gerganov ggerganov@gmail.com Website: htWed, 21 Oct 2026 11:58:23 GMTllama-cpphttps://github.com/mastra-ai/mastra/releases/tag/mastracode%400.48.0https://github.com/mastra-ai/mastra/releases/tag/mastracode%400.48.0mastracode@0.48.0Wed, 21 Oct 2026 11:57:33 GMTmastrahttps://github.com/mastra-ai/mastra/releases/tag/mastra%401.35.0https://github.com/mastra-ai/mastra/releases/tag/mastra%401.35.0mastra@1.35.0Wed, 21 Oct 2026 11:57:33 GMTmastrahttps://github.com/mastra-ai/mastra/releases/tag/%40mastra%2Fweaviate%400.1.3https://github.com/mastra-ai/mastra/releases/tag/%40mastra%2Fweaviate%400.1.3@mastra/weaviate@0.1.3Wed, 21 Oct 2026 11:57:33 GMTmastrahttps://github.com/BerriAI/litellm/releases/tag/v1.107.0https://github.com/BerriAI/litellm/releases/tag/v1.107.0Verify Docker Image Signature All LiteLLM Docker images are signed with cosign . Every release is signed with the same key introduced in commit 0112e53 . Verify using the pinned commit hash (recommended): A commit hash is cryptographically immutable, so this is the strongest way Wed, 21 Oct 2026 11:50:24 GMTlitellmhttps://github.com/langfuse/langfuse/releases/tag/v4.67.0https://github.com/langfuse/langfuse/releases/tag/v4.67.0What's Changed Features feat(skills): simplify CLI install and import setup by @tobi12345 in #19128 feat(web): add name search and gateway key filter to organization API keys by @wochinge in #19108 Fixes fix(slack): marketplace review – link buttons and private channel visibilityWed, 21 Oct 2026 11:31:17 GMTlangfusehttps://github.com/BerriAI/litellm/releases/tag/v1.106.2https://github.com/BerriAI/litellm/releases/tag/v1.106.2Verify Docker Image Signature All LiteLLM Docker images are signed with cosign . Every release is signed with the same key introduced in commit 0112e53 . Verify using the pinned commit hash (recommended): A commit hash is cryptographically immutable, so this is the strongest way Wed, 21 Oct 2026 10:50:35 GMTlitellmhttps://github.com/BerriAI/litellm/releases/tag/v1.105.3https://github.com/BerriAI/litellm/releases/tag/v1.105.3Verify Docker Image Signature All LiteLLM Docker images are signed with cosign . Every release is signed with the same key introduced in commit 0112e53 . Verify using the pinned commit hash (recommended): A commit hash is cryptographically immutable, so this is the strongest way Wed, 21 Oct 2026 10:00:46 GMTlitellmhttps://github.com/n8n-io/n8n/releases/tag/n8n%402.44.5https://github.com/n8n-io/n8n/releases/tag/n8n%402.44.52.44.5 (2026-10-21) Bug Fixes core: Key queue job outcomes by job on shutdown resume ( #41404 ) ( 1f074d4 ) editor: Load credential types on startup ( #41405 ) ( 6ff3198 )Wed, 21 Oct 2026 07:30:18 GMTn8nhttps://github.com/n8n-io/n8n/releases/tag/stablehttps://github.com/n8n-io/n8n/releases/tag/stable2.44.5 (2026-10-21) Bug Fixes core: Key queue job outcomes by job on shutdown resume ( #41404 ) ( 1f074d4 ) editor: Load credential types on startup ( #41405 ) ( 6ff3198 )Wed, 21 Oct 2026 07:27:36 GMTn8nhttps://github.com/firecrawl/firecrawl/releases/tag/v2.11.580https://github.com/firecrawl/firecrawl/releases/tag/v2.11.580Release API image v2.11.580Wed, 21 Oct 2026 04:52:43 GMTfirecrawlhttps://github.com/anomalyco/opencode/releases/tag/github-v1.2.27https://github.com/anomalyco/opencode/releases/tag/github-v1.2.27chore(function): remove exchange_github_app_token_with_pat route ( #57 …Wed, 21 Oct 2026 04:08:35 GMTopencodehttps://github.com/firecrawl/firecrawl/releases/tag/v2.11.579https://github.com/firecrawl/firecrawl/releases/tag/v2.11.579Release API image v2.11.579Wed, 21 Oct 2026 04:04:45 GMTfirecrawlhttps://github.com/ollama/ollama/releases/tag/v0.40.7https://github.com/ollama/ollama/releases/tag/v0.40.7What's Changed pplx-decider is a 27B decision model from Perplexity, fine-tuned from Qwen3.8-27B Continuing an OpenAI-compatible Responses conversation no longer fails when switching between models with and without thinking, or when the history contains another provider's compactWed, 21 Oct 2026 04:01:19 GMTollamahttps://github.com/anomalyco/opencode/releases/tag/v2.1.2https://github.com/anomalyco/opencode/releases/tag/v2.1.2release: v2.1.2Wed, 21 Oct 2026 03:41:04 GMTopencodehttps://github.com/firecrawl/firecrawl/releases/tag/v2.11.578https://github.com/firecrawl/firecrawl/releases/tag/v2.11.578Release API image v2.11.578Wed, 21 Oct 2026 03:37:21 GMTfirecrawlhttps://github.com/anthropics/claude-code/releases/tag/v2.1.306https://github.com/anthropics/claude-code/releases/tag/v2.1.306What's changed Fixed claude -p inside cloud sessions reporting "Not logged in" when started without the session's environment variables Fixed an API error being replaced by "prompt is too long", and an unneeded compaction, in sessions with many MCP toolsWed, 21 Oct 2026 03:16:31 GMTclaude-codehttps://github.com/Unstructured-IO/unstructured/releases/tag/0.27.32https://github.com/Unstructured-IO/unstructured/releases/tag/0.27.32What's Changed fix: label numbered headings without making them list items, and skip section-break paragraphs by @qued in #4538 fix(partition): read spooled uploads in place instead of copying them by @CyMule in #4419 perf(pdf): bound OCR upload and rendered-page memory by @CyMulTue, 20 Oct 2026 23:13:41 GMTunstructuredhttps://github.com/Comfy-Org/ComfyUI/releases/tag/v0.42.3https://github.com/Comfy-Org/ComfyUI/releases/tag/v0.42.3ComfyUI v0.42.3Tue, 20 Oct 2026 22:44:30 GMTcomfyuihttps://github.com/docling-project/docling/releases/tag/v2.141.0https://github.com/docling-project/docling/releases/tag/v2.141.0Feature Add nvidia/Nemotron-3.5-Parse as supported VLM model ( #4897 ) ( 2d1d814 ) cli: Generic engine options for docling convert ( #4729 ) ( f57160e ) service: Allow artifact downloads from trusted private origins ( #4892 ) ( 8df5b6e ) forms: Map form fields with a geometric soTue, 20 Oct 2026 21:27:15 GMTdoclinghttps://github.com/anthropics/claude-code/releases/tag/v2.1.305https://github.com/anthropics/claude-code/releases/tag/v2.1.305What's changed Added sandbox.allowNestedUserNamespaces so sandboxed commands on Linux can create user namespaces (Chromium, rootless podman); this lets them get around the sandbox's file protections Added CLAUDE_CODE_RETRY_WATCHDOG_USAGE_LIMIT_RECHECK_MS so unattended retry mode Tue, 20 Oct 2026 20:46:50 GMTclaude-codehttps://github.com/Comfy-Org/ComfyUI/releases/tag/v0.42.2https://github.com/Comfy-Org/ComfyUI/releases/tag/v0.42.2ComfyUI v0.42.2Tue, 20 Oct 2026 20:34:13 GMTcomfyuihttps://vercel.com/changelog/websocket-support-is-now-generally-availablehttps://vercel.com/changelog/websocket-support-is-now-generally-availableWebSocket support in Vercel Functions (used for multiplayer apps, chat streaming and AI Gateway's realtime models) is now GA.Tue, 20 Oct 2026 19:56:01 GMTvercelhttps://github.com/langfuse/langfuse/releases/tag/v4.66.0https://github.com/langfuse/langfuse/releases/tag/v4.66.0What's Changed Features feat(topics): one gate for Topics visibility by @LysanderKie in #19091 Fixes fix(sessions): Don't allow expanding empty reasoning messages by @bezbac in #19096 fix(web): disable Owner role selection without ownership permission by @marksalpeter in #19092 fTue, 20 Oct 2026 17:31:25 GMTlangfusehttps://github.com/anomalyco/opencode/releases/tag/v2.1.1https://github.com/anomalyco/opencode/releases/tag/v2.1.1release: v2.1.1Tue, 20 Oct 2026 13:36:07 GMTopencodehttps://github.com/langfuse/langfuse/releases/tag/v4.65.0https://github.com/langfuse/langfuse/releases/tag/v4.65.0What's Changed Features feat(topics): non-blocking clustering for large numbers by @nimarb in #18964 feat(topics): name topics by what their facet asks about by @LysanderKie in #19008 feat(routes): add recovery option for V4 experience and update sidebar toggle logic by @marliessTue, 20 Oct 2026 13:35:55 GMTlangfusehttps://github.com/browserbase/stagehand/releases/tag/%40browserbasehq%2Fstagehand%404.3.0https://github.com/browserbase/stagehand/releases/tag/%40browserbasehq%2Fstagehand%404.3.0@browserbasehq/stagehand@4.3.0Tue, 20 Oct 2026 00:36:14 GMTstagehandhttps://github.com/browserbase/stagehand/releases/tag/stagehand-python%404.3.0https://github.com/browserbase/stagehand/releases/tag/stagehand-python%404.3.0This PR was opened by the Changesets release GitHub action. When you're ready to do a release, you can merge this and the packages will be published to npm automatically. If you're not ready to do a release yet, that's fine, whenever you add more changesets to main, this PR will Tue, 20 Oct 2026 00:24:52 GMTstagehandhttps://github.com/browserbase/stagehand/releases/tag/packages%2Fsdk-go%2Fv4.3.0https://github.com/browserbase/stagehand/releases/tag/packages%2Fsdk-go%2Fv4.3.0This PR was opened by the Changesets release GitHub action. When you're ready to do a release, you can merge this and the packages will be published to npm automatically. If you're not ready to do a release yet, that's fine, whenever you add more changesets to main, this PR will Tue, 20 Oct 2026 00:24:52 GMTstagehandhttps://developers.deepgram.com/changelog/2026/10/20https://developers.deepgram.com/changelog/2026/10/20Moves Flux STT to a new Engine runtime (needs new .dgv2 Flux model files), adds mid-stream keyterm updates for Nova-3, makes Flux Force End Turn always available, and improves formatting and media handling.Tue, 20 Oct 2026 00:00:00 GMTdeepgramhttps://github.com/Unstructured-IO/unstructured/releases/tag/0.27.30https://github.com/Unstructured-IO/unstructured/releases/tag/0.27.30What's Changed fix(email): keep non-ASCII sender and recipient names readable by @usmanmateen in #4543 Full Changelog : 0.27.29...0.27.30Mon, 19 Oct 2026 23:08:14 GMTunstructuredhttps://github.com/anthropics/claude-code/releases/tag/v2.1.304https://github.com/anthropics/claude-code/releases/tag/v2.1.304What's changed Added a transcript line when a plugin that runs first skips one of your mod's hooks, naming the event and the plugin Added decider: "person" to a mod's tool.check answer: only the permission dialog or a host's callback can approve the ask, and it is refused with thMon, 19 Oct 2026 21:34:39 GMTclaude-codehttps://github.com/Unstructured-IO/unstructured/releases/tag/0.27.29https://github.com/Unstructured-IO/unstructured/releases/tag/0.27.29What's Changed fix(xlsx): read .xls files whose Workbook stream ends in padding by @usmanmateen in #4545 New Contributors @usmanmateen made their first contribution in #4545 Full Changelog : 0.27.28...0.27.29Mon, 19 Oct 2026 20:39:51 GMTunstructuredhttps://github.com/langchain-ai/langgraph/releases/tag/cli%3D%3D0.4.37https://github.com/langchain-ai/langgraph/releases/tag/cli%3D%3D0.4.37Changes since cli==0.4.36 release(cli): 0.4.37 ( #9351 ) fix(cli): stop printing subprocess stdin in verbose output ( #9350 ) Merge PR #9349 fix(cli): default deploy to serverless ( #9333 ) chore(deps): bump the minor-and-patch group in /libs/cli/js-examples with 9 updates ( #915Mon, 19 Oct 2026 14:06:22 GMTlanggraphhttps://github.com/pipecat-ai/pipecat/releases/tag/v1.13.0https://github.com/pipecat-ai/pipecat/releases/tag/v1.13.0Added Added support for clients sending files and images to the bot via the new RTVI send-file message, incorporating them into the LLM context. New frames: FileRawFrame , InputFileRawFrame , and UserFileRawFrame , carrying a file payload as inline base64 bytes or as a URL. New LMon, 19 Oct 2026 06:35:05 GMTpipecathttps://github.com/sgl-project/sglang/releases/tag/v0.5.22https://github.com/sgl-project/sglang/releases/tag/v0.5.22Highlights 950 PRs from 245 contributors. New models Model Type Cookbook Clef / Clef Flash LLM / VLM link PPLX-Decider v1 / v1.1 LLM / VLM link EmbeddingGemma 2 LLM / VLM link LLaDA2.2 LLM / VLM HunyuanImage-3.0-Instruct Diffusion link Kandinsky 6 / Kandinsky 6 SR Diffusion link Mon, 19 Oct 2026 05:55:53 GMTsglanghttps://github.com/Comfy-Org/ComfyUI/releases/tag/v0.42.1https://github.com/Comfy-Org/ComfyUI/releases/tag/v0.42.1Fix some crashing issues a few users experienced when running the MOSS image template workflow.Mon, 19 Oct 2026 01:33:54 GMTcomfyuihttps://developers.deepgram.com/changelog/2026/10/19https://developers.deepgram.com/changelog/2026/10/19Every Configure message on a /v1/listen stream now gets exactly one ConfigureSuccess or ConfigureFailure reply. Python SDK v7.13.0, JavaScript SDK v5.15.0 and Java SDK v0.11.1 add mid-stream Nova-3 keyterm updates; Deepgram CLI v0.3.3 adds dg speak --play.Mon, 19 Oct 2026 00:00:00 GMTdeepgramhttps://exa.ai/changelog#october-19-2026https://exa.ai/changelog#october-19-2026New type: "lite" on /search for latency- and cost-sensitive work ($4 per 1k requests with up to 10 results, cache-only content). fast, instant and deep-reasoning become legacy. Responses add warnings and notices arrays; exa-py and exa-js 2.26.0 surface them.Mon, 19 Oct 2026 00:00:00 GMTexahttps://github.blog/changelog/2026-10-19-copilot-model-policy-api-for-enterprise-orgs-and-teamshttps://github.blog/changelog/2026-10-19-copilot-model-policy-api-for-enterprise-orgs-and-teamsGET and PATCH REST endpoints for Copilot model policies are now generally available at enterprise, enterprise-team, agent-app and organization level.Mon, 19 Oct 2026 00:00:00 GMTgithub-copilothttps://github.blog/changelog/2026-10-19-github-copilot-models-deprecated-on-october-19-2026https://github.blog/changelog/2026-10-19-github-copilot-models-deprecated-on-october-19-2026Gemini 3.7 Flash, GPT-5.5, GPT-5.4, GPT-5.4 mini, GPT-5 mini and Grok 4.5 are deprecated across most Copilot experiences. Suggested replacements: Gemini 3.8 Flash, GPT-6.1 Sol, GPT-6 Luna and Grok 4.7.Mon, 19 Oct 2026 00:00:00 GMTgithub-copilothttps://docs.coderabbit.ai/changelog#gitlab-migration-csv-inputhttps://docs.coderabbit.ai/changelog#gitlab-migration-csv-inputProject webhooks can now be migrated from headerless CSV lists of project or group IDs, full paths or GitLab URLs with the existing --project and --group flags.Mon, 19 Oct 2026 00:00:00 GMTcoderabbithttps://github.com/vllm-project/vllm/releases/tag/v0.32.0https://github.com/vllm-project/vllm/releases/tag/v0.32.0v0.32.0 Highlights This release features 814 commits from 363 contributors (121 new)! Structured decisions and generic parsers : the new POST /v1/systemone endpoint answers choice , noul and score questions over a structured state ( #59299 ), is served by default ( #60677 , #6069Sun, 18 Oct 2026 10:23:45 GMTvllmhttps://github.com/chroma-core/chroma/releases/tag/latesthttps://github.com/chroma-core/chroma/releases/tag/latestVersion: 1.5.10.dev324 Git ref: refs/heads/main Build Date: 2026-10-17T22:33 PIP Package: chroma-1.5.10.dev324.tar.gz Github Container Registry Image: :1.5.10.dev324 DockerHub Image: :1.5.10.dev324Sat, 17 Oct 2026 22:33:49 GMTchromahttps://github.com/ollama/ollama/releases/tag/v0.40.6https://github.com/ollama/ollama/releases/tag/v0.40.6What's Changed win: prevent orphaned runners by @dhiltgen in #18500 gguf: add the q2_0 tensor and file types by @jmorganca in #18967 Launch clef in Ollama's cloud by @jberg5 in #19036 use identifiers for --ask mode question keys by @jberg5 in #19041 Full Changelog : v0.40.5...v0.Sat, 17 Oct 2026 15:29:04 GMTollamahttps://github.com/cline/cline/releases/tag/sdk%2Fsdk%2Fv0.0.95https://github.com/cline/cline/releases/tag/sdk%2Fsdk%2Fv0.0.95OpenAI-compatible requests for GPT-6 models ( gpt-6-sol , gpt-6.1-sol , gpt-6-1-sol , and later gpt-5 through gpt-9 ids) send max_completion_tokens instead of max_tokens , so OpenAI and Azure no longer reject them. Azure's gpt-35-turbo keeps max_tokens The hub daemon records whicSat, 17 Oct 2026 03:00:13 GMTclinehttps://github.com/cline/cline/releases/tag/sdk%2Fllms%2Fv0.0.95https://github.com/cline/cline/releases/tag/sdk%2Fllms%2Fv0.0.95@cline/llms@0.0.95Sat, 17 Oct 2026 02:59:54 GMTclinehttps://github.com/cline/cline/releases/tag/sdk%2Fshared%2Fv0.0.95https://github.com/cline/cline/releases/tag/sdk%2Fshared%2Fv0.0.95@cline/shared@0.0.95Sat, 17 Oct 2026 02:59:52 GMTclinehttps://github.com/langchain-ai/langgraph/releases/tag/1.2.16https://github.com/langchain-ai/langgraph/releases/tag/1.2.16Changes since 1.2.15 release(langgraph): 1.2.16 ( #9334 ) fix(langgraph): let other tasks finish when a resume value is invalid ( #9265 ) fix(langgraph): apply a resumed Command's update once ( #9322 ) fix(langgraph): stop a subgraph from rerunning its finished tasks on resume ( Fri, 16 Oct 2026 22:11:28 GMTlanggraphhttps://github.com/langchain-ai/langgraph/releases/tag/sdk%3D%3D0.4.8https://github.com/langchain-ai/langgraph/releases/tag/sdk%3D%3D0.4.8Changes since sdk==0.4.7 release(langgraph): 1.2.16 ( #9334 ) release(sdk-py): 0.4.8 ( #9335 ) fix(sdk-py): surface thread stream errors and stop retrying a rejected stream ( #9317 ) release(prebuilt): 1.1.1 ( #9316 )Fri, 16 Oct 2026 22:09:01 GMTlanggraphhttps://github.com/confident-ai/deepeval/releases/tag/typescript-v0.9.23https://github.com/confident-ai/deepeval/releases/tag/typescript-v0.9.23Deepeval setup scriptFri, 16 Oct 2026 20:18:37 GMTdeepevalhttps://github.com/confident-ai/deepeval/releases/tag/setup-v0.1.0https://github.com/confident-ai/deepeval/releases/tag/setup-v0.1.0What's Changed docs: Add missing params to Generate Goldens From Documents guide by @AbdoAlshoki2 in #3311 . by @penguine-ip in #3337 feat(persona): metadata and noise url by @tanayvaswani in #3333 fix(models): add Claude Opus 5.5 to the Anthropic registry (AmazonBedrockModel uncFri, 16 Oct 2026 20:00:06 GMTdeepevalhttps://github.com/confident-ai/deepeval/releases/tag/typescript-v0.9.22https://github.com/confident-ai/deepeval/releases/tag/typescript-v0.9.22new releaseFri, 16 Oct 2026 19:59:23 GMTdeepevalhttps://github.com/convex/convex-backend/releases/tag/precompiled-2026-10-10-72ce13chttps://github.com/convex/convex-backend/releases/tag/precompiled-2026-10-10-72ce13cusage: aggregate database bandwidth (#59420) Remove the redundant empty-batch check in audit-log delivery; both Firehose transports already handle empty batches. Allow database bandwidth events to omit `request_id` for summaries spanning multiple requests. Individual execution evFri, 16 Oct 2026 16:22:41 GMTconvexhttps://github.com/axolotl-ai-cloud/axolotl/releases/tag/v0.21.0https://github.com/axolotl-ai-cloud/axolotl/releases/tag/v0.21.0Axolotl v0.21.0 Release Notes 59 commits since v0.20.0 (September 30, 2026). This release is headlined by Reflection's Beam, a 501B-parameter MoE you can QLoRA-tune on a single node. We also added native diffusion language model training, image datasets for DPO, KTO, and GRPO, anFri, 16 Oct 2026 13:52:11 GMTaxolotlhttps://github.com/supabase/supabase/releases/tag/self-hosted%2Fv0.8.3https://github.com/supabase/supabase/releases/tag/self-hosted%2Fv0.8.3Self-hosted Supabase v0.8.3: Security fixes for Storage and Realtime,…Fri, 16 Oct 2026 11:43:13 GMTsupabasehttps://github.com/docling-project/docling/releases/tag/v2.140.0https://github.com/docling-project/docling/releases/tag/v2.140.0Feature asr: Prefetch the native Distil-Whisper presets with docling-tools ( #4825 ) ( 483cf13 ) asciidoc: Parse bold, italic and monospace in paragraphs ( #4824 ) ( fd24b9e ) asr: Transcribe with a remote OpenAI-compatible server ( #4853 ) ( b246052 ) Fix odf: Move hidden ODT seFri, 16 Oct 2026 09:00:49 GMTdoclinghttps://github.com/browser-use/browser-use/releases/tag/0.13.13https://github.com/browser-use/browser-use/releases/tag/0.13.13This release includes the OpenAI browser-tool driver and fixes two integration issues. Handle temporary empty Cloud tab URLs during navigation. Keep screenshots, crops, and pointer coordinates aligned on scaled displays. Provide 18 default browser tools and six optional actions tThu, 15 Oct 2026 23:50:45 GMTbrowser-usehttps://github.com/browser-use/browser-use/releases/tag/0.13.12https://github.com/browser-use/browser-use/releases/tag/0.13.12Add OpenAI browser tools backed by Browser Use Actor. Import BrowserUse from browser_use.integrations.openai . Install browser-use[openai-tools] to enable OpenAI Python 3.31+. Use local browsers, Browser Use Cloud, or an existing connected BrowserSession. Support the 18 default bThu, 15 Oct 2026 23:37:58 GMTbrowser-usehttps://github.com/openai/codex/releases/tag/rust-v0.163.0https://github.com/openai/codex/releases/tag/rust-v0.163.0New Features Open /config to search preferences across categories, change supported settings immediately, and see which changes require a restart. ( #52990 , #53429 ) Send a follow-up while Codex is responding to steer it immediately. Interrupting a turn also stops active Code MoThu, 15 Oct 2026 21:33:06 GMTopenai-codexhttps://github.com/ollama/ollama/releases/tag/v0.40.5https://github.com/ollama/ollama/releases/tag/v0.40.5NVIDIA Nemotron 3.5 Super NVIDIA Nemotron 3.5 Super is a vision-language model for coding, reasoning, tool use, and document understanding. It pairs a 120B-parameter hybrid Mamba-2/Transformer mixture-of-experts backbone (12B active per token) with a C-RADIO vision encoder, and sThu, 15 Oct 2026 19:18:39 GMTollamahttps://github.com/livekit/agents/releases/tag/livekit-agents%401.8.8https://github.com/livekit/agents/releases/tag/livekit-agents%401.8.8GPT-Live through LiveKit Inference You can now run GPT-Live with your LiveKit credentials instead of a separate OpenAI key. inference.DuplexModel handles its protocol, and you can adopt it directly with a model string: session = AgentSession ( llm = "openai/gpt-live-1" ) New provThu, 15 Oct 2026 06:10:17 GMTlivekit-agentshttps://github.com/invoke-ai/InvokeAI/releases/tag/win-arm64-wheels-v1https://github.com/invoke-ai/InvokeAI/releases/tag/win-arm64-wheels-v1Self-built win_arm64 wheels consumed by pyproject.toml's invoke-win-arm64-wheels index. Byte-identical copies of the assets first published in the archived invoke-ai/InvokeAI-7.Thu, 15 Oct 2026 03:11:55 GMTinvokeaihttps://github.com/openai/openai-agents-python/releases/tag/v0.24.0https://github.com/openai/openai-agents-python/releases/tag/v0.24.00.24.0 (2026-10-15) Features expose MCP annotations on converted tools ( #5356 ) ( 75813f8 ) support run-wide tool exception formatting ( #5326 ) ( 8fd9a7c ) Bug Fixes batch OpenAI conversation session writes within the API limit ( #5334 ) ( 26345c1 ) defer automatic compaction fThu, 15 Oct 2026 02:37:43 GMTopenai-agents-sdkhttps://github.com/vllm-project/vllm/releases/tag/proto-v0.6.0https://github.com/vllm-project/vllm/releases/tag/proto-v0.6.0vllm-proto 0.6.0Wed, 14 Oct 2026 23:57:02 GMTvllmhttps://github.com/docling-project/docling/releases/tag/v2.139.0https://github.com/docling-project/docling/releases/tag/v2.139.0Feature Expose max_concurrent_results on ThreadedDoclingParseBackendOptions ( #4804 ) ( ce5621b ) Expose per-page confidence breakdown in ConfidenceScores ( #4233 ) ( 5bdabb9 ) progress: Add conversion progress bars and an event callback ( #4560 ) ( fdb6321 ) iwork: Read picturesWed, 14 Oct 2026 12:18:27 GMTdoclinghttps://github.com/unclecode/crawl4ai/releases/tag/v0.9.5https://github.com/unclecode/crawl4ai/releases/tag/v0.9.5🎉 Crawl4AI v0.9.5 Released! 📦 Installation PyPI: pip install crawl4ai==0.9.5 Docker: docker pull unclecode/crawl4ai:0.9.5 docker pull unclecode/crawl4ai:latest Note: Docker images are being built and will be available shortly. Check the Docker Release workflow for build status.Wed, 14 Oct 2026 10:25:39 GMTcrawl4aihttps://github.com/unclecode/crawl4ai/releases/tag/docker-rebuild-v0.9.5https://github.com/unclecode/crawl4ai/releases/tag/docker-rebuild-v0.9.5Merge main into develop: cloud launch wordingWed, 14 Oct 2026 10:23:58 GMTcrawl4aihttps://github.com/livekit/agents/releases/tag/livekit-agents%401.8.7https://github.com/livekit/agents/releases/tag/livekit-agents%401.8.7Realtime E2E latency You can now measure realtime replies from the last detected user speech to first assistant playback when you configure local VAD. Read the latency from assistant message metrics: @ session . on ( "conversation_item_added" ) def on_message ( event ): if event Wed, 14 Oct 2026 04:29:17 GMTlivekit-agentshttps://github.com/activepieces/activepieces/releases/tag/0.92.3https://github.com/activepieces/activepieces/releases/tag/0.92.3🐞 Bug Fixes fix(engine): force reinstall piece when require fails ( #16106 ) @MrChaker fix(worker): installing pieces no longer depends on the worker reaching the public URL ( #16037 ) @AhmadTash 📌 Upgrading a self-hosted instance? Review the Breaking Changes page for any actioMon, 12 Oct 2026 18:43:34 GMTactivepieceshttps://github.com/convex/convex-backend/releases/tag/precompiled-2026-10-10-bb95347https://github.com/convex/convex-backend/releases/tag/precompiled-2026-10-10-bb95347Add search query arg to logs page (#59453) Should make it easy to link to a page w/ search params (filter params, and component/function/level selection) GitOrigin-RevId: d6828ecf19f1fb6ed9515c870fe509994ae4e3aaSat, 10 Oct 2026 00:38:51 GMTconvexhttps://github.com/livekit/agents/releases/tag/livekit-agents%401.8.6https://github.com/livekit/agents/releases/tag/livekit-agents%401.8.6What's Changed fix(sarvam): close the streams the STT collects by @dhruvladia-sarvam in #7058 fix(aws): keep realtime session active after stream restart by @uuzzrm in #6811 fix(telemetry): format a loop stall report without reading source files by @zdurm in #7616 update deps by Fri, 09 Oct 2026 22:17:26 GMTlivekit-agentshttps://github.com/convex/convex-backend/releases/tag/precompiled-2026-10-09-deb555ahttps://github.com/convex/convex-backend/releases/tag/precompiled-2026-10-09-deb555aRelease ai-sdk-provider 0.3.0 (#59425) Bump `@convex-dev/ai-sdk-provider` to `0.3.0` for the stable release of voice support and opt-in ZDR routing. GitOrigin-RevId: 31541635c471ba945dab5f34d62862dce0ff71fcFri, 09 Oct 2026 16:13:48 GMTconvexhttps://github.com/qdrant/qdrant/releases/tag/v1.19.0https://github.com/qdrant/qdrant/releases/tag/v1.19.0Change log Features ☝️ https://github.com/qdrant/qdrant/milestone/50 - TurboQuant 4-bit as a datatype of primary vector storage. Only store 4-bit quantized vectors and spare disk space on original vectors. [ docs ] #9669 , #9684 , #9950 - Unify definition of memory usage strategyFri, 09 Oct 2026 10:25:54 GMTqdranthttps://github.com/vllm-project/vllm/releases/tag/proto-v0.5.0https://github.com/vllm-project/vllm/releases/tag/proto-v0.5.0Release vllm-proto 0.5.0Fri, 09 Oct 2026 03:11:40 GMTvllmhttps://github.com/promptfoo/promptfoo/releases/tag/code-scan-action-0.2.2https://github.com/promptfoo/promptfoo/releases/tag/code-scan-action-0.2.20.2.2 (2026-10-08) Bug Fixes code-scan: mint OIDC after CLI installation ( #11467 ) ( 9a404c8 )Thu, 08 Oct 2026 15:46:03 GMTpromptfoohttps://github.com/promptfoo/promptfoo/releases/tag/0.124.1https://github.com/promptfoo/promptfoo/releases/tag/0.124.10.124.1 (2026-10-08) Features assertions: record cost without a threshold ( #11451 ) ( 037f1f6 ) providers: add OpenAI Decisions support ( #11454 ) ( 2dbf3ec ) providers: expose catalog text-token pricing ( #11452 ) ( 3ecce2e ) Bug Fixes assertions: preserve async expression gradThu, 08 Oct 2026 14:23:27 GMTpromptfoo