Home / Shipped today

Shipped today

Stable releases from tools we rank, last 14 days. Every item links to the official release. RSS feed

Stable GitHub releases of tracked tools (pre-releases filtered). Date is the <updated> field of the GitHub releases Atom feed. Updated Oct 19, 2026. At most 3 per tool.

Oct 19, 2026

Firecrawl · v2.11.548
Release API image v2.11.548
Claude Code · v2.1.304
What'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 th
llama.cpp · b11742
hybrid dna tokenizer fixes ( #30825 ) vocab : support Carbon-14B HybridDNA tokenizer Carbon-14B auto-tags untagged DNA runs and uses N-padded tail tokens (tail=pad_n). Also strip whitespace inside tags and handle a stray like the HF tokenizer. convert : fix special tokens and mer
Unstructured · 0.27.29
What'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...0.27.29
llama.cpp · b11741
opencl: add cooperative-k dp4a gemm for q4_0, q4_k and q6_k ( #30741 ) opencl: add cooperative-k dp4a GEMM for q4_0 opencl: add cooperative-k dp4a GEMM for q4_k opencl: add cooperative-k dp4a GEMM for q6_k opencl: fix alignment Website: https://llama.app Attestations: https://git
llama.cpp · b11740
opencl: add get_rows for i32, cpy for i32 and f32 ( #30756 ) opencl: support i32 GET_ROWS opencl: support i32 and f32 copy conversions Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/attestations/57528236 macOS/iOS: macOS Apple Silicon (arm64) macOS
Firecrawl · v2.11.547
Release API image v2.11.547
Firecrawl · v2.11.546
Release API image v2.11.546
Langfuse · v4.64.0
What's Changed Features feat(trace): Move messages tab and reuse session timeline by @bezbac in #18953 feat(api-keys): enable project API key permissions by default by @marksalpeter in #18950 feat: rename Langfuse Assistant to Halo by @Lotte-Verheyden in #18898 feat(api-keys): sh
LangGraph · langgraph-cli==0.4.37
Changes 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 ( #915
n8n · n8n@2.44.4
2.44.4 (2026-10-19) Bug Fixes core: Reuse connections in the log streaming webhook destination ( #41281 ) ( b79afe2 ) core: Turn off expression spans by default ( #41309 ) ( 6965193 )
n8n · n8n@2.43.8
2.43.8 (2026-10-19) Bug Fixes core: Turn off expression spans by default ( #41310 ) ( e88162a )
n8n · stable
2.43.8 (2026-10-19) Bug Fixes core: Turn off expression spans by default ( #41310 ) ( e88162a )
Pipecat · v1.13.0
Added 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 L
SGLang · v0.5.22
Highlights 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
ComfyUI · v0.42.1
Fix some crashing issues a few users experienced when running the MOSS image template workflow.

Oct 18, 2026

Claude Code · v2.1.303
What's changed Bug fixes and reliability improvements
ComfyUI · v0.42.0
What's Changed [Partner Nodes] feat(Krea): add Krea 3 Image node by @bigcat88 in #17067 Update workflow templates to v0.11.85 by @comfyui-wiki in #17068 Use FLUX 3 Image in node display labels by @MaanavD in #17075 Better documentation about the comfy quant format. by @comfyanony
vLLM · v0.32.0
v0.32.0 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 , #6069
LiteLLM · v1.106.0
Verify 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

Oct 17, 2026

Chroma · Latest
Version: 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.dev324
LiteLLM · v1.103.5
Verify 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
Ollama · v0.40.6
What'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.
ComfyUI · v0.41.3
ComfyUI v0.41.3
Cline · SDK v0.0.95
OpenAI-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 whic
Cline · sdk/llms/v0.0.95
@cline/llms@0.0.95
Cline · sdk/shared/v0.0.95
@cline/shared@0.0.95

Oct 16, 2026

OpenCode · v2.0.32
release: v2.0.32
LangGraph · langgraph==1.2.16
Changes 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 (
LangGraph · langgraph-sdk==0.4.8
Changes 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 )
Unstructured · 0.27.28
What's Changed feat: opt-in langdetect-rs backend for partition language detection by @kholdrex in #4547 fix(html): keep text inside unrecognized elements by @huuufu in #4497 fix(html): keep table captions in Table text and text_as_html by @Sriram-PR in #4553 perf(docx): resolve
Claude Code · v2.1.302
What's changed Added CLAUDE_CODE_AUTO_MODE_MODEL to choose the model the client-side auto mode classifier uses; it does not change the model for server-side auto mode Added an Idle compaction row to /config for turning idle compaction off or back on; the notices shown after an id
DeepEval · typescript-v0.9.23: Merge pull request #3474 from confident-ai/deepeval-setup-script
Deepeval setup script
DeepEval · Setup wizard 0.1.0
What'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 unc
DeepEval · typescript-v0.9.22
new release
Convex · Precompiled 2026-10-10-72ce13c
usage: 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 ev
Langfuse · v4.63.0
Features feat(web): link personal feature previews from organization settings by @nimarb in #18887 feat(traces): integrate PostHog analytics for IO parser comparison tracking by @marliessophie in #18896 feat(skills): improve browsing and editing UX by @tobi12345 in #18940 feat(ai
Axolotl · v0.21.0
Axolotl 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, an
OpenCode · v2.0.31
release: v2.0.31
OpenCode · v1.19.3
Core Improvements Configure model reasoning options to generate reasoning variants, replacing catalog variants while preserving your configured variant overrides.
Supabase · self-hosted/v0.8.3
Self-hosted Supabase v0.8.3: Security fixes for Storage and Realtime,…
Langfuse · v4.62.0
What's Changed ci: run ai-gateway e2e tests on ai-gateway crate changes by @wochinge in #18915 ci(deps): let the security maintainer remediate Rust Dependabot alerts by @wochinge in #18885 fix(web): tracing page header and search row spacing by @nmtrang29 in #18352 feat(ai-gatewa
Docling · v2.140.0
Feature 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 se

Oct 15, 2026

Browser Use · 0.13.13: OpenAI browser tools and Cloud fixes
This 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 t
Browser Use · 0.13.12: OpenAI browser tools
Add 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 b
Mastra · mastracode@0.47.1
mastracode@0.47.1
Mastra · mastra@1.34.2
mastra@1.34.2
Mastra · @mastra/weaviate@0.1.2
@mastra/weaviate@0.1.2
OpenAI Codex · 0.163.0
New 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 Mo
Ollama · v0.40.5
NVIDIA 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 s
LiveKit Agents · livekit-agents@1.8.8
GPT-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 prov
InvokeAI · Windows ARM64 wheels (win-arm64-wheels-v1)
Self-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.
OpenAI Agents SDK · v0.24.0
0.24.0 (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 f
Unstructured · 0.27.27
What's Changed fix(json): preserve rehydrated element identity by @rohan-patnaik in #4504 fix: keep overlap category names from element 100 on by @jipeng6036-del in #4529 fix: normalize None text to empty string in Text elements by @TrueFurina in #4463 fix: stop silently returnin

Oct 14, 2026

vLLM · proto-v0.6.0
vllm-proto 0.6.0
Stagehand · browse@0.11.2
browse@0.11.2
Docling · v2.139.0
Feature 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 pictures
Crawl4AI · Release 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.
Crawl4AI · docker-rebuild-v0.9.5
Merge main into develop: cloud launch wording
LiteLLM · v1.105.1
Verify 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
LiveKit Agents · livekit-agents@1.8.7
Realtime 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
Ollama · v0.40.4
Decision models in the CLI Decision models can now be used directly from the terminal with ollama run — ask yes/no questions, classify text into labels, or score against a rubric: ollama run nimble --ask "Is a refund being requested?" < ticket.txt What's Changed Kev decision m

Oct 12, 2026

Activepieces · 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 actio
Docling · v2.138.0
Feature iwork: Read pictures placed on a Numbers sheet ( #4700 ) ( 88fbc05 ) Fix docx: Emit textboxes in table cells once inside their cell ( #4757 ) ( 6bc5770 ) jats: Parse ordered lists and list item labels ( #4770 ) ( ad2e0e1 ) mlx: Pass the sampling temperature as temperature

Oct 10, 2026

Stagehand · stagehand-python@4.2.1: Release Stagehand packages (#3160)
This 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
Stagehand · packages/sdk-go/v4.2.1: Release Stagehand packages (#3160)
This 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
Convex · Precompiled 2026-10-10-bb95347
Add 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: d6828ecf19f1fb6ed9515c870fe509994ae4e3aa

Oct 9, 2026

LiveKit Agents · livekit-agents@1.8.6
What'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
Convex · Precompiled 2026-10-09-deb555a
Release 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: 31541635c471ba945dab5f34d62862dce0ff71fc
Qdrant · v1.19.0
Change 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 strategy
vLLM · proto-v0.5.0
Release vllm-proto 0.5.0

Oct 8, 2026

Promptfoo · code-scan-action: 0.2.2
0.2.2 (2026-10-08) Bug Fixes code-scan: mint OIDC after CLI installation ( #11467 ) ( 9a404c8 )
Promptfoo · 0.124.1
0.124.1 (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 grad

Oct 7, 2026

Activepieces · 0.92.2
🧹 Maintenance chore(engine): evaluate template-expression scripts with the legacy sandbox ( #16189 ) @MrChaker chore(engine): use the legacy sandbox for the shared script session too ( #16190 ) @MrChaker This routes {{ }} template-expression evaluation through the legacy sandbox
Browser Use · 0.13.11
This release includes Browser Use toolsets for Claude, available at browser_use.integrations.toolsets_for_claude . The integration requires an Anthropic SDK release containing anthropic.tools.browser ; the current public Anthropic SDK does not yet include it. Documentation: https

Oct 6, 2026

Promptfoo · 0.124.0
0.124.0 (2026-10-06) ⚠ BREAKING CHANGES providers: remove hosted ChatKit provider ( #11434 ) providers: make WatsonX SDKs opt-in ( #11359 ) deps: make Langfuse SDK opt-in ( #11351 ) deps: make Slack SDK opt-in ( #11349 ) providers: make OpenAI Agents an explicit install ( #11177