# Blog Read about our latest announcements. [All](https://devin.ai/blog)[Product](https://devin.ai/blog/category/product)[Enterprise](https://devin.ai/blog/category/enterprise)[Changelist](https://devin.ai/blog/category/changelist)[Company](https://devin.ai/blog/category/company)[Industry](https://devin.ai/blog/category/industry)[Case Studies](https://devin.ai/blog/category/case-studies) [![](https://devin.ai/_next/image?url=%2Fimages%2Fblog%2Fdevins-slack-etiquette%2Fat-devin-texture-1.jpg\&w=3840\&q=75\&dpl=dpl_GYjHJaEAH5tMYx9wAxhM4XyY5rRt)](https://devin.ai/blog/devins-slack-etiquette) [Product](https://devin.ai/blog/devins-slack-etiquette) ## [Devin's Slack Etiquette](https://devin.ai/blog/devins-slack-etiquette) [How we refined Devin's behavior in Slack over the past year, across prompts, Devin's harness, and our usage of the Slack API.](https://devin.ai/blog/devins-slack-etiquette) [August 20, 20266 min read](https://devin.ai/blog/devins-slack-etiquette) [![](https://devin.ai/_next/image?url=%2Fimages%2Fblog%2Fgpt-5-6-sol-promo%2Fcover.png\&w=3840\&q=75\&dpl=dpl_GYjHJaEAH5tMYx9wAxhM4XyY5rRt)](https://devin.ai/blog/gpt-5-6-sol-promo) [Product](https://devin.ai/blog/gpt-5-6-sol-promo) ## [GPT-5.6 Sol: 70% off in Devin](https://devin.ai/blog/gpt-5-6-sol-promo) [GPT-5.6 Sol is available at a 70% discount on Devin Desktop and Devin CLI through October 3, 2026.](https://devin.ai/blog/gpt-5-6-sol-promo) [August 18, 20261 min read](https://devin.ai/blog/gpt-5-6-sol-promo) [![](https://devin.ai/_next/image?url=%2Fimages%2Fblog%2Fgemini-3.7-flash%2Fcover.png\&w=3840\&q=75\&dpl=dpl_GYjHJaEAH5tMYx9wAxhM4XyY5rRt)](https://devin.ai/blog/gemini-37-flash) [Product](https://devin.ai/blog/gemini-37-flash) ## [Gemini 3.7 Flash is now available in Devin](https://devin.ai/blog/gemini-37-flash) [Gemini 3.7 Flash is now live in Devin Desktop and Devin CLI, reaching Claude Sonnet 5-level performance at less than half the cost. Available at a 50% discount for two weeks.](https://devin.ai/blog/gemini-37-flash) [August 13, 20261 min read](https://devin.ai/blog/gemini-37-flash) [![](https://devin.ai/_next/image?url=%2Fimages%2Fblog%2Fgrok-4-6%2Fcover.png\&w=3840\&q=75\&dpl=dpl_GYjHJaEAH5tMYx9wAxhM4XyY5rRt)](https://devin.ai/blog/grok-4-6) [Product](https://devin.ai/blog/grok-4-6) ## [Grok 4.6 is now available in Devin](https://devin.ai/blog/grok-4-6) [Grok 4.6 is now live in Devin Desktop and the Devin CLI, marking a significant improvement over Grok 4.5 on FrontierCode 1.1.](https://devin.ai/blog/grok-4-6) [August 12, 20261 min read](https://devin.ai/blog/grok-4-6) [![](https://devin.ai/_next/image?url=%2Fimages%2Fblog%2Fintroducing-pr-stacks%2Fpr-stacks-thumbnail.png\&w=3840\&q=75\&dpl=dpl_GYjHJaEAH5tMYx9wAxhM4XyY5rRt)](https://devin.ai/blog/introducing-pr-stacks) [Product](https://devin.ai/blog/introducing-pr-stacks) ## [Introducing Stacked PRs in Devin](https://devin.ai/blog/introducing-pr-stacks) [Devin now breaks large tasks into stacks of small, independently reviewable PRs — handling rebasing, conflict resolution, and CI verification for you.](https://devin.ai/blog/introducing-pr-stacks) [July 30, 20266 min read](https://devin.ai/blog/introducing-pr-stacks) [![](https://devin.ai/_next/image?url=%2Fimages%2Fblog%2Fkimi-k3%2Fcover.png\&w=3840\&q=75\&dpl=dpl_GYjHJaEAH5tMYx9wAxhM4XyY5rRt)](https://devin.ai/blog/kimi-k3) [Product](https://devin.ai/blog/kimi-k3) ## [Kimi K3 is now available in Devin](https://devin.ai/blog/kimi-k3) [Kimi K3 is now live in Devin Desktop and Devin CLI. Optimized for long horizon agentic coding, it approaches frontier-level FrontierCode 1.1 performance.](https://devin.ai/blog/kimi-k3) [July 27, 20261 min read](https://devin.ai/blog/kimi-k3) [![](https://devin.ai/_next/image?url=%2Fimages%2Fblog%2Fclaude-opus-5%2Fcover.png\&w=3840\&q=75\&dpl=dpl_GYjHJaEAH5tMYx9wAxhM4XyY5rRt)](https://devin.ai/blog/claude-opus-5) [Product](https://devin.ai/blog/claude-opus-5) ## [Claude Opus 5 is now available in Devin](https://devin.ai/blog/claude-opus-5) [Claude Opus 5 is now live in Devin Desktop, Devin CLI, and Devin Cloud's mode mix, approaching Fable-level performance at half the cost.](https://devin.ai/blog/claude-opus-5) [July 23, 20261 min read](https://devin.ai/blog/claude-opus-5) [![](https://devin.ai/_next/image?url=%2Fimages%2Fblog%2Fintroducing-devin-outposts%2Foutposts-thumbnail.png\&w=3840\&q=75\&dpl=dpl_GYjHJaEAH5tMYx9wAxhM4XyY5rRt)](https://devin.ai/blog/introducing-devin-outposts) [Product](https://devin.ai/blog/introducing-devin-outposts) ## [Introducing Devin Outposts](https://devin.ai/blog/introducing-devin-outposts) [Give Devin the ability to run on any machine.](https://devin.ai/blog/introducing-devin-outposts) [July 21, 20265 min read](https://devin.ai/blog/introducing-devin-outposts) [![](https://devin.ai/_next/image?url=%2Fimages%2Fblog%2Fgpt-5-6%2Fcover.png\&w=3840\&q=75\&dpl=dpl_GYjHJaEAH5tMYx9wAxhM4XyY5rRt)](https://devin.ai/blog/gpt-5-6) [Product](https://devin.ai/blog/gpt-5-6) ## [GPT-5.6 models are now available in Devin](https://devin.ai/blog/gpt-5-6) [GPT-5.6 models are now live in Devin. These models stand out for combining top-tier coding-agent performance with very strong token/cost efficiency.](https://devin.ai/blog/gpt-5-6) [July 9, 20261 min read](https://devin.ai/blog/gpt-5-6) [![](https://devin.ai/_next/image?url=%2Fimages%2Fblog%2Fclaude-fable-5-available-in-devin%2Fcover.png\&w=3840\&q=75\&dpl=dpl_GYjHJaEAH5tMYx9wAxhM4XyY5rRt)](https://devin.ai/blog/claude-fable-5-available-in-devin) [Product](https://devin.ai/blog/claude-fable-5-available-in-devin) ## [Update on Claude Fable 5 in Devin](https://devin.ai/blog/claude-fable-5-available-in-devin) [Claude Fable 5 is now available in Devin Cloud’s Ultra agent, Devin Desktop, and Devin CLI.](https://devin.ai/blog/claude-fable-5-available-in-devin) [July 1, 20261 min read](https://devin.ai/blog/claude-fable-5-available-in-devin) [![](https://devin.ai/_next/image?url=%2Fimages%2Fblog%2Fagentic-map-reduce%2Fmapreduce-illustration.jpg\&w=3840\&q=75\&dpl=dpl_GYjHJaEAH5tMYx9wAxhM4XyY5rRt)](https://devin.ai/blog/agentic-map-reduce) [Product](https://devin.ai/blog/agentic-map-reduce) ## [Agentic MapReduce](https://devin.ai/blog/agentic-map-reduce) [A new architecture from Cognition for scaling full-codebase reasoning with distributed agents](https://devin.ai/blog/agentic-map-reduce) [July 1, 20269 min read](https://devin.ai/blog/agentic-map-reduce) [![](https://devin.ai/_next/image?url=%2Fimages%2Fblog%2Fsecurity-swarm-eval%2Fsecurity-blog-illustration.jpg\&w=3840\&q=75\&dpl=dpl_GYjHJaEAH5tMYx9wAxhM4XyY5rRt)](https://devin.ai/blog/security-swarm-eval) [Product](https://devin.ai/blog/security-swarm-eval) ## [Evaluating Security Swarm](https://devin.ai/blog/security-swarm-eval) [How we built an eval of real, recent vulnerabilities to measure Security Swarm — and why it finds more bugs at lower cost than any tool we tested.](https://devin.ai/blog/security-swarm-eval) [July 1, 20268 min read](https://devin.ai/blog/security-swarm-eval)