How many hours slip away each week on repetitive tasks? Why not automate them? Mark your calendar for October 10th at Shack 15 in San Francisco. Join us to build agent in just 4 hours with @coderabbitai, @ClerkDev @neondatabase and @Vapi_AI! Compete for a chance to win up to $15K and Cline credits.

What's the daily chore that eats away your time? Imagine a tool that handles it for you! On October 10th in SF we are joining with the @coderabbitai team to host a hackathon at Shack 15 to automate the boring stuff. 4 hours. $35k in prizes. #SFTechWeek

We released a patch this week (v3.32.3) which includes GLM-4.6. The model hits 94.9% success rate on diff edits in our testing. That's within 1.3 points of Sonnet 4.5 at 10% the cost.

The Cline team is heading to Mexico City and we want you to come build with us! October 21-22 at PLX'25, spend two days solving real industry challenges alongside engineers, designers, and PMs from across LatAm.

👀 x.com/juanpflores...

Pull up to Cline & Wine 🍷 at SF Tech Week! Sip, vibe, and network with the SF tech scene. Catch the latest from Saoud ( @sdrzn) and Yaz El-Baba ( @YazanElBaba) at Emergence Capital, and snack on some bites. 👉🏼Oct 8, 5-8pm Don't miss it!

GLM-4.6 from @Zai_org is available Cline. 200K context (up from 131k), and completes tasks with 15% fewer tokens than GLM-4.5. >48.6% win rate against frontier models, making it one of the most capable open-source models. Available in Cline & the GLM subscription.

Live in Cline: Claude Sonnet 4.5 What's different from Sonnet 4: > more terse, less narration > maintains state across sessions via context files > enhanced capability across long tasks 200k or 1M context window & the same $3/$15 pricing (increased > 200k tokens)

Workflow automation in 3 steps: 1. Complete any repetitive task in Cline 2. Type: /create-new-workflow (prompt linkedin below) 3. Next time: /your-workflow Your 30-minute process becomes a 30-second command. Try it with your next PR review or test suite run.

stealth model update: 5x more context `code-supernova-1-million` (access via cline provider) > 1 million token context window (up from 200k) > same multi-modal support > same free access during alpha <the supernova is intensifying>

🔍🦙 x.com/ollama/stat...

Voice Mode (powered by Whisper) isn't just faster than typing. It's about complete intent transfer. Try it in Plan mode for rapid collaboration.

Just shipped Cline v3.31: (1) Voice Mode, (2) YOLO Mode, & (3) a sleeker Task Header Available in all VS Code IDEs & forks, plus JetBrains. <thread>

Just shipped @jetbrains v1.01, which includes a patch to support the Rider IDE.

See you this week at @nerdearla 2025! Cline will be present as a sponsor at the largest free tech conference in Latin America 🇦🇷 📅 September 23-27 📍 Buenos Aires + Online 🎯 2 sessions on AI development

<<free stealth model just dropped 🥷>> `code-supernova`: - 200k context window - multi-modal (i.e. image inputs) - "built for agentic coding" - completely free during alpha access via the cline provider: `cline:cline/code-supernova` what are we building this weekend??

We rebuilt Cline from the ground up for @jetbrains. cline-core runs as a headless process, communicating through gRPC. True native integration with JetBrains APIs, not emulation. This architecture is our foundation for bringing Cline to any IDE.

1. Model agnostic. 2. Inference agnostic. & now, 3. Platform agnostic. Cline for JetBrains is here. (install it below)

Module 2 of AI Coding University just dropped! The LLM landscape exploded this year. Claude 4 Sonnet, GPT-5, local models. More choices than ever, but which one fits your workflow? That's exactly what Module 2: LLM Fundamentals covers. 🧵

Cline v3.28: Two big wins today 🧵 1. Free Grok extended (was ending today) 2. GPT-5 runs better in Cline

Got some free time this weekend? We just launched our first module on AI Coding University 📚 where we teach core skills developers benefit from to use AI-powered development assistants.

Tired of AI tools causing more friction than gains? We just launched our first module on AI Coding University 📚 where we teach core skills developers benefit from to use AI-powered development assistants.

ATTENTION NEW YORKERS 🚨 Join us for the world's fastest hackathon, with a 2 hour timeline only possible when powered by @CerebrasSystems inference x Cline. Even better, run the chance to win 𝙇𝙖𝙗𝙪𝙗𝙪𝙨 (yes!), 𝘿𝙪𝙗𝙖𝙞 𝙘𝙝𝙤𝙘𝙤𝙡𝙖𝙩𝙚, and $1000 in cash prizes! 👹🍫✨

Last call, SF! 🚨 This Saturday at Shack15, 𝙗𝙪𝙞𝙡𝙙 𝘼𝙄 𝙖𝙜𝙚𝙣𝙩𝙨 𝙩𝙝𝙖𝙩 𝙙𝙤𝙣'𝙩 𝙟𝙪𝙨𝙩 𝙘𝙝𝙖𝙩, 𝙗𝙪𝙩 𝙙𝙤. Connect to APIs, automate workflows, make decisions, and take actions. Your personal AI employee that works 24/7. $20k in prizes. Weekend-long hacking. Free food. Co-hosted by Cline and @coderabbitai 🤖 🐇

It's a wrap! Thank you so much to everyone who attended the Lightspeed hackathon in Nashville TN. For 3 days, participants pushed their physical, intellectual, and creative boundaries. What made this event truly special was watching the winning teams leverage Cline to build incredible projects.

New stealth Sonoma Sky/Dusk models with 2M token context windows (!) recently appeared on the @vercel AI Gateway and @OpenRouterAI. We've been tracking their performance in Cline since they released in stealth. 🧵 (you can try them for free btw!)

Live in Cline, Kimi-K2-0905 via the @FireworksAI_HQ provider.

.@Kimi_Moonshot just released Kimi-K2-0905 and here's the scoop: > 2x context window (131k to 256k) > trained for improved tool use, esp in agent harnesses like Cline > it's markedly better at frontend Try it now in Cline 🌔

Join us & @coderabbitai for an all-weekend hackathon at Shack15 in SF! Signup below 👇

We crossed 50k GitHub stars ⭐️

🚀 Nashville builders: from Sept 5-7, join LIGHTSPEED at The Lighthouse, a 48-hour hackathon to ship real MVPs. $50 in Cline credits for every participant + $5K in prizes. RSVP today: luma.com/jtt7j965

Three ways to code for free just landed in Cline v3.26.6. Cloud speed with @grok Code Fast 1, local privacy via @LMStudio, or generous daily limits via the Qwen Code provider --> pick your path. 🧵

DeepSeek V3.1 dropped last week and is showing a 9.9% diff edit failure rate in real-world usage. That's higher than Qwen 3 Coder (6.1%) but still solid for an open-source model for an early release. What do you think? How's it been in Cline so far?

So many incredible builders in NYC 🗽 Shoutout to @cerebral_valley for having us! x.com/cerebral_va...

The latest from @deepseek_ai, deepseek-chat-v3.1, is live in Cline. It boasts a SWEBench score of 66%, closely rivaling Sonnet 4's 72.7%. It's context window is 164k and is more affordable at $0.56/$1.68 per M tokens.

.@GroqInc has introduced prompt caching for the SOTA open-source coding model, Kimi-k2. What does this mean for you? Significantly reduced costs with Kimi-k2 on its fastest provider. Check the price: left side without prompt caching, right side with prompt caching.

New stealth model in Cline: "Sonic" Designed for coding & free to use, because your usage helps improve the model while it's in alpha.

LLMs need focus, because attention isn't enough. (thread on the Focus Chain, Cline's link to persistent context)

/extension-release.md

More choice for Cline users: you can now run models from DeepSeek, Moonshot AI, Qwen, and Meta via @basetenco, who is focused on delivering reliable inference to the SOTA coding models. Pick the inference provider that works best for your workflow.

GPT-5 has been hovering around a 7% diff edit failure rate since its release to Cline last Thursday. How have you liked GPT-5 so far in Cline?

Live in Cline via @huggingface: gpt-oss-120b & gpt-oss-20b > gpt-oss-120b is built for production-grade applications with high reasoning capabilities > gpt-oss-20b is for lower latency needs and specialized local use cases Also accessible via the Cline & OpenRouter providers

We've noticed a significant focus on improving the tool-calling success rate in open-source models -- a traditionally weak area for them. x.com/cline/statu...

Live in Cline: GLM 4.5 from @Zai_org, an agent-focused, open-source model designed for coding and tool use. Key Specs: > 128k context window, 96k max output > $0.60/$2.20 per million tokens > 90.6% tool-calling success rate (highest among tested models)

Qwen3 Coder has a 5.32% diff edit failure rate in Cline, based on real-world data. It's right alongside Claude Sonnet 4 and Kimi K2 as an excellent model in performing diff edits. Not bad for an open-source model at $0.30/$1.20 per million input/output tokens 👀

How are we feeling about @Alibaba_Qwen's qwen3-coder? (FYI, its diff edit failure rate, 5 days in, is 5.75%; on par with Sonnet 4 and Kimi K2)

The worst part of @CerebrasSystems inference? You don't time to make an espresso while Cline codes. Join our hackathon next Saturday to explore the new paradigm: instant inference. $5k in prizes → RSVP below

> 2500 tokens per second > 40x faster than Sonnet-4 > 4 trillion transistor wafers (19x Nvidia B200) These are impressive numbers, but nothing beats a live demo. 1 million free tokens per day of lightning-fast qwen3-32b and qwen3-235b from @CerebrasSystems in Cline. Link below.

Qwen3-Coder is now available in Cline 🧵 New 480B parameter model with 35B active parameters. > 256K context length natively > comparable performance to Claude Sonnet 4 > SoTA among open source models

Kimi K2 outperforms Gemini 2.5 Pro and matches Sonnet-4 in real-world diff edit failure rates.