Introducing SWE-2 Flash, our new fast & efficient coding model.
On leading evals, it outperforms all American open-weight models, even ones 3x larger.
To balance transparency and cyber risks, we are building a program to securely share the weights with partners & researchers.
you can now edit Devin Apps visually: tweak CSS properties, change out design tokens, and annotate changes for your agent visually
these design tools are coming to your local apps in Desktop soon too :)
Devin Apps are powered by memory snapshotting VMs. That means this goes far beyond your normal "app builder":
1. Nontechnical builders can do more than landing pages now
2. Enterprise internal apps are "secure by design"
3. PR preview links go beyond "just frontend"
Devin Apps is your new home for:
▪️ Preview deployments of full working apps
▪️ Analytics tools connected to production databases
▪️ Internal tooling and microsites
Now available in Devin.
Get started: https://t.co/SPMExihcEB
Introducing: Devin Apps
Private, secure, sharable web applications running directly on Devin’s cloud computer.
Devin Apps are designed to power your most ambitious internal tools, preview deployments, and extensions to Devin itself.
have we seen all there is of coding agents? probably not.
today's @DevinAI deli event also doubled as a housewarming for @cognition's singapore office! japan fdes also came by to share their perspectives on where they think deployment is going 🧵
I hosted an invite-only "Devin's Dinner" last night for startup founders & engineers & other interesting people(tm) in NYC.
It was such a blast that I'm thinking I'll maybe do another one next month as well!
DM me if you're interested.
Claude Mythos 5.2 is now available in Devin’s managed security scans.
We are giving $100M+ in credits to help companies find and fix their most critical vulnerabilities with Claude Mythos 5.2. Eligibility criteria and conditions apply.
On our internal cyber benchmark, Mythos 5.2 scores 73.5, a substantial step up from Mythos 5.1’s 56.9.
Gemini 4 Carbon is now available in Devin Desktop and CLI.
On FrontierCode 2.0, Carbon scores 40.0% – a substantial step up from Gemini 4 Argon’s 26.3% – making it the best performing Google model on the benchmark.
Claude Fable 5.2 is now available in Devin Desktop and Devin CLI.
On FrontierCode 2.0, Fable 5.2 takes the #1 spot at a score of 49.1.
Fable 5.2 also introduces a significant price improvement from Fable 5.1: at low reasoning effort, it surpasses its predecessor’s score at a third of the cost.
A few weeks ago we shared our research on evals with @cognition & @braintrust. Now meet the team that built it:
@gopalkgoel1, @XuBriGuy, and @AksharSarvesh share how our production traces become evals automatically.
If problems like this excite you, we’re hiring: https://t.co/qs7neUoy7b
We've shipped a lot of features to Devin and Devin Desktop, let's dive in to 20 things you might not know Devin can do.
Starting with Local Hands: Devin Cloud can run commands on your machine.
Simple Natural Language Environment Setup: In one prompt Devin configures its own cloud environment and builds a reusable snapshot for all of your future sessions, including your environment variables.