## 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 models are now supported by the MLX runner on Apple silicon
* `ollama launch` can now be used with usage credits without upgrading to a paid plan
* `ollama launch codex` now includes locally installed Ollama models in its model picker. Use `/model` within the Codex CLI to swap models.
* Improved inference performance by reusing verified manifests and already loaded models
* Added support for multimodal embeddings through llama-server
* Fixed loading multimodal embedding models with small context windows
* Improved Gemma 4 prefill performance on NVIDIA GPUs with the MLX runner
* MLX runner now clips long Laya states instead of rejecting them
* Improved speculative decodin
## 🐞 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](https://www.activepieces.com/docs/install/reference/breaking-changes) page for any actions required before you upgrade.
## Thanks ❤️
@MrChaker, @AhmadTash
## What's Changed
### Features
* feat(ee): refuse FIPS mode with storage options that compute MD5 by @Steffen911 in https://github.com/langfuse/langfuse/pull/18443
* feat(worker): add optional startup jitter before queue registration by @niklassemmler in https://github.com/langfuse/langfuse/pull/18525
* feat(auth): hash passwords with PBKDF2 in FIPS mode by @Steffen911 in https://github.com/langfuse/langfuse/pull/18444
* feat(web): restore optional Cloudflare Turnstile captcha on credential auth by @maxdeichmann in https://github.com/langfuse/langfuse/pull/18580
* feat(web): carry the typed email over to the password reset form by @dergeraetbot in https://github.com/langfuse/langfuse/pull/18585
* feat(gateway): add API key expiration by @wochinge in https://github.com/langfuse/langfuse/pull/18561
* feat(evals): measure score persistence outcomes by @wochinge in https://github.com/langfuse/langfuse/pull/18600
* feat(skills): add CLI installation guidance by @tobi12345 in ht
## 1.27.0 (2026-10-12)
<!-- Release notes generated using configuration in .github/release.yml at main -->
## What's Changed
### Features
* feat(verify-openhands): drive the mock-LLM e2e behaviors on DeepSeek and map the gaps by @enyst in https://github.com/OpenHands/OpenHands/pull/18191
* feat(verify-openhands): record videos of the browser with `browser record` by @enyst in https://github.com/OpenHands/OpenHands/pull/18211
* feat(acp): add OpenCode and Pi as built-in ACP agents on local backends by @Placidoe in https://github.com/OpenHands/OpenHands/pull/17758
* feat: Add splash page mode toggle by @malhotra5 in https://github.com/OpenHands/OpenHands/pull/17199
* feat: add automate setup split panel by @malhotra5 in https://github.com/OpenHands/OpenHands/pull/17200
* feat: add agent-assisted automation form updates by @malhotra5 in https://github.com/OpenHands/OpenHands/pull/17201
* feat: add persisted automation draft management UI by @malhotra5 in https://github.com/OpenHands/Ope
## Features
- feat(library): How Do You Build a HubSpot AI Agent for Lead Enrichment and Follow-Up? (#8905)
- feat(library): How to Build a Pipedrive CRM AI Agent for Deal Follow-Ups (#8903)
- feat(projects): move Project membership to the workspace column (#8830)
## Improvements
- perf(tables): seek keyset row pages instead of scanning from the first key (#8911)
- perf(auth): stop blocking requests on best-effort last-used timestamp writes (#8910)
- perf(knowledge): lock the knowledge base row FOR NO KEY UPDATE on document writes (#8909)
- perf(files): resolve name references through an index instead of listing the workspace (#8908)
- perf(search): fetch Lucid page regions concurrently on a live read (#8890)
- perf(search): search up to eight live accounts at once (#8891)
## Bug Fixes
- fix(search): return actionable live read errors and drop unreadable Confluence matches (#8888)
- fix(file-parsers): cap untrusted number formats and make time-format check linear (#8893)
- fix(rea
## What's Changed
* `ollama launch claude` no longer shows a warning about Claude connectors by @hoyyeva in https://github.com/ollama/ollama/pull/18878
* `ollama launch codex` no longer warns about service tiers by @hoyyeva in https://github.com/ollama/ollama/pull/18879
* Paused background model upgrade for embedding and other request performance, by @jmorganca in https://github.com/ollama/ollama/pull/18908
**Full Changelog**: https://github.com/ollama/ollama/compare/v0.40.2...v0.40.3
## Features
- feat(changelog): workspace changelog of releases with markdown bodies and the Sim changelog tool (#8859)
- feat(library): AI Agents for Asana and monday.com Task Automation: Platform and Workflow Guide (#8847)
- feat(sim-cli): publish which CLI commands Mothership is refused, from the route admission rule (#8838)
- feat(shopify): expand operations and refresh expiring credentials (#8833)
- feat(checkr): add Checkr integration with tools, block, and webhook triggers (#8824)
- feat(tests): workflow tests as a workspace resource (#8773)
- feat(library): Enterprise AI Agent Compliance and Deployment Guide: HIPAA, GDPR, Data Residency, On-Prem, and VPC (#8822)
- feat(changelog): publish curated product updates (#8787)
- feat(analytics): attribute sign-ups and demo requests to their acquisition source (#8815)
- feat(dashboards): add empty-state graphic for the dashboard page (#8719)
- feat(oracledb): add Oracle Database integration (#7378)
- feat(library): 6 Best AI Agent Mark
## What's Changed
### Features
- Add XPU to OpenCLIP provider device options
- Add deepinfra as an OpenAI-compatible provider
- Record why an evaluation stopped before its verdict
- Record what the run app's Deploy met before an attempt
- Implement per-run cost and time tracking in crewai eval --models without repeated progress lines
### Bug Fixes
- Fix issue where stopSequences were sent to OpenAI GPT-6/GPT-5.6/gpt-oss
- Preserve GitHub loader source attribution
- Fix SQLite flow initialization contention
### Documentation
- Add snapshot and changelog for v1.15.26
## Contributors
@StoyanD, @ats3v, @joaomdmoura, @kimnamu, @lorenzejay, @mhbuehler, @zouyonghe
## Features
* feat(web): design-system CodeSection replaces CodeView by @nmtrang29 in https://github.com/langfuse/langfuse/pull/18462
* feat(sessions): Add syntax highlighting to session tool previews by @bezbac in https://github.com/langfuse/langfuse/pull/18472
* feat(billing): send spend alerts for ClickHouse-billed organizations by @Steffen911 in https://github.com/langfuse/langfuse/pull/17615
* feat(web): ask new users what they will use Langfuse for by @AbdallahAHO in https://github.com/langfuse/langfuse/pull/18356
* feat(topics): add adaptive map exploration and compact workspace by @nkabardin in https://github.com/langfuse/langfuse/pull/18492
* feat(evals): revamp decision model results by @wochinge in https://github.com/langfuse/langfuse/pull/18523
* feat(tables): let rows use a custom height by @nkabardin in https://github.com/langfuse/langfuse/pull/18445
## Fixes
* fix(web): design-system Timeline with dotted faded connectors by @nmtrang29 in https://github.com/lan
## [2.43.3](https://github.com/n8n-io/n8n/compare/n8n@2.43.2...n8n@2.43.3) (2026-10-09)
### Features
* **core:** Block create/update of workflows with deprecated nodes ([#40706](https://github.com/n8n-io/n8n/issues/40706)) ([3c4f6e3](https://github.com/n8n-io/n8n/commit/3c4f6e33934adf29d2c76e7723256fa5b69b9ab9))
* **Google Gemini Node:** Show Nano Banana 2.1 and 2 Lite in the image model list ([#40655](https://github.com/n8n-io/n8n/issues/40655)) ([1830ccf](https://github.com/n8n-io/n8n/commit/1830ccfcda55fc948197fbb88ac7438211a39940))
<!-- This is an auto-generated description by cubic. -->
<a href="https://cubic.dev/pr/n8n-io/n8n/pull/40736?utm_source=github" target="_blank" rel="noopener noreferrer" data-no-image-dialog="true"><picture><source media="(prefers-color-scheme: dark)" srcset="https://www.cubic.dev/buttons/review-in-cubic-dark.svg"><source media="(prefers-color-scheme: light)" srcset="https://www.cubic.dev/buttons/review-in-cubic-light.svg"><img alt="View guided diff"
## [2.42.6](https://github.com/n8n-io/n8n/compare/n8n@2.42.5...n8n@2.42.6) (2026-10-09)
### Features
* **core:** Block create/update of workflows with deprecated nodes ([#40707](https://github.com/n8n-io/n8n/issues/40707)) ([ff1a24c](https://github.com/n8n-io/n8n/commit/ff1a24c72f360ce9d047f17eef08e0732c829f4b))
* **Google Gemini Node:** Show Nano Banana 2.1 and 2 Lite in the image model list ([#40656](https://github.com/n8n-io/n8n/issues/40656)) ([8dfc695](https://github.com/n8n-io/n8n/commit/8dfc695c8c6a9f799a1780ac76444770c174f8a1))
<!-- This is an auto-generated description by cubic. -->
<a href="https://cubic.dev/pr/n8n-io/n8n/pull/40735?utm_source=github" target="_blank" rel="noopener noreferrer" data-no-image-dialog="true"><picture><source media="(prefers-color-scheme: dark)" srcset="https://www.cubic.dev/buttons/review-in-cubic-dark.svg"><source media="(prefers-color-scheme: light)" srcset="https://www.cubic.dev/buttons/review-in-cubic-light.svg"><img alt="View guided diff"
## What's Changed
### Bug Fixes
- Fix ownership retention of rwlock when acquisition is interrupted.
- Resolve issue with documentation site source URLs not being preserved.
- Fix task output display in log-tasks-outputs to show actual output.
### Documentation
- Update snapshot and changelog for version v1.15.25.
## Contributors
@Velmet44, @Vidit-Ostwal, @joaomdmoura, @leilei3167, @lorenzejay, @lux-liang
## 1.26.0 (2026-10-08)
<!-- Release notes generated using configuration in .github/release.yml at main -->
## What's Changed
### Features
* feat(apps): add authenticated backend view helper by @neubig in https://github.com/OpenHands/OpenHands/pull/17656
* feat(chat): render empty-response corrective nudge as an informational note by @errmakov in https://github.com/OpenHands/OpenHands/pull/17864
* feat: verify-openhands skill with control-openhands CLI and the full feature map by @enyst in https://github.com/OpenHands/OpenHands/pull/17961
* feat(verify-openhands): daily pass support and CLI fixes by @enyst in https://github.com/OpenHands/OpenHands/pull/18093
* feat(verify-openhands): map 20 more behaviors, with live recipes by @enyst in https://github.com/OpenHands/OpenHands/pull/18105
* feat(backends): explain a suspended workspace and offer a switch by @hieptl in https://github.com/OpenHands/OpenHands/pull/17988
* feat(onboarding): show the Super Admin setup guide on enterprise back
## Model upgrades
Models downloaded with earlier versions of Ollama are upgraded in the background the first time you run them, for better performance and compatibility when running on llama.cpp.
To make downgrading safe, Ollama keeps the original copy as a backup, so upgraded models are temporarily kept on disk. A future release will remove these backups automatically.
To remove backed up models now (requires `jq`):
```shell
curl -s localhost:11434/api/tags | jq -r '.models[].name' | while read -r m; do
curl -s localhost:11434/api/show -d "{\"model\":\"$m\"}" | jq -r \
'select(any(.manifests[]?; .runner == "llamacpp")) | .manifests[] | select(.runner == "ggml") | .digest'
done | sort -u | xargs -n1 ollama rm
```
This only deletes the backups. If you later downgrade to a version older than 0.40, you'll need to re-pull those models.
**Other changes**
* `ollama list` no longer shows duplicate entries for upgraded models by @dhiltgen in https://github.com/oll
## What's Changed
### Features
* feat(docker): create zero vuln base images and add fips mode by @Steffen911 in https://github.com/langfuse/langfuse/pull/18009
* feat(topics): rework the summary system prompt and default facets by @LysanderKie in https://github.com/langfuse/langfuse/pull/17935
* feat(web): show ClickHouse organization id in org debug information by @Steffen911 in https://github.com/langfuse/langfuse/pull/18390
* feat(topics): summarize all facets of a trace in one call by @LysanderKie in https://github.com/langfuse/langfuse/pull/18013
* feat(skills): import skills from local files and ZIPs by @tobi12345 in https://github.com/langfuse/langfuse/pull/18378
* feat(skills): add rendered file previews with raw source access by @tobi12345 in https://github.com/langfuse/langfuse/pull/18396
* feat(evals): support OpenAI decision models by @wochinge in https://github.com/langfuse/langfuse/pull/18389
* feat(analytics): include error ids in toast events by @nkabardin
## Features
- feat(library): GitHub MCP Server for AI Agents: Builders, Security, and Setup (#8794)
- feat(library): Best Slack MCP Agent Builders: 7 Options Compared (#8793)
- feat(library): How to Use the GitHub MCP Server With AI Agents: Builders, Security, and Setup (#8791)
## Bug Fixes
- fix(db): compare tables by schema-qualified name in the migration audit (#8801)
- fix(deps): patch sharp librsvg CVE and MCP SDK OAuth issuer binding (#8798)
- fix(db): refuse migrations whose ownership statements only pass outside production (#8796)
- fix(build): keep zod's default error messages in production builds (#8797)
## Other Changes
- docs(library): update govern-ai-agents-multiple-teams-enterprise-workspace (#8795)
- docs(library): update best-conversational-ai-platforms-crm-helpdesk (#8792)
- docs(library): update what-is-human-in-the-loop-in-ai-agents (#8790)
- ci: bump the actions group across 4 directories with 19 updates (#8789)
- docs(library): update open-source-ai-agent-pla