[HN] Comments

Preliminary benchmark results for the new OpenAI o1 models.

An Architect model describes how to solve the coding problem, and an Editor model translates that into file edits. This Architect/Editor approach produces SOTA benchmark results.

Open source LLMs are becoming very powerful, but pay attention to how you (or your provider) are serving the model. It can affect code editing skill.

QwQ is reasoning model like o1, and needs to be used as an architect with another model as editor.

o1 scores the top result on aider's new multi-language, more challenging coding benchmark.

Reliably packaging & distributing python CLI tools is hard. Aider uses uv in novel ways to make it easy to install the aider CLI, its dependencies and python 3.12. All in an isolated env.

R1+Sonnet has set a new SOTA on the aider polyglot benchmark. At 14X less cost compared to o1.

DeepSeek's API has been experiencing reliability issues. Here are alternative providers you can use.

The $6.32 benchmark cost reported for Gemini 2.5 Pro Preview 03-25 was incorrect.

Benchmark results for Qwen3 models using the Aider polyglot coding benchmark.

Insights from Codeium into our products, tools, and community.

Copilot++ now has chunked streaming (currently in Beta)! It will surface edits faster in smaller chunks. To enable it, click the settings gear and enable "Chunked Streaming" under Features > Copilot++

Cursor Tab (previously called Copilot++) defaults to chunked streaming. This build also includes several Cursor Tab speedups. More to come in future builds! Concurrent composers support, composer cont

We have a new chat UX! Excited for you to try it out and share your thoughts. Composer is now default-on and available to all Pro/Business users by hitting cmd+I. We've added Composer Projects (beta),

*This update fixes the following security issue: * *CVE-2024-45599* *.* Cursor Tab now auto-imports symbols in Python files! We've also significantly improved the Cursor Tab stability. Composer Notepa

Composer history lets you access previous composer sessions after restart. You can also edit and resubmit from previous messages within a session. We have made slight improvements to Debug with AI and

Composer UI in the sidebar with inline diffs Early version of an agent in composer that can pick its own context and use the terminal Generation of git commit messages File pill recommendations in cha

Agent sees terminal exit codes, can run commands in the background, and commands are now editable Agent reads linter errors to automatically fix issues With Yolo Mode, the agent can auto-run terminal

.cursor/rules: Users can write several repository-level rules to disk in the .cursor/rules directory. The Agent will automatically choose which rule to follow. Deepseek models: Deepseek R1 and Deepsee

Fixes (3) **Crashes**: Enhanced stability through memory and performance improvements, with additional crash prevention systems in place. **MCP**: Improved reliability and quality of life enhancements

Patches (11) 0.47.1: Improved performance, added back play button to apply code blocks 0.47.2: Cursor Tab accepts work with single-line selections 0.47.3: Fixes an issue with tool call errors on file

Other (6) Improved indexing performance for teams sharing codebases The vertical sidebar option will temporarily be unavailable for users who opted into Early Access Projects with MCP servers will now

Automated and improved rules You can now generate rules directly from a conversation using the `/Generate Cursor Rules` command. This is useful when you want to capture the existing context of a conve

Introducing unified request-based pricing, Max Mode for all top models, and Background Agent for parallel task execution. Plus, improved context management with `@folders` support, refreshed Inline Ed

Cursor 1.0 is here! This release brings Bugbot for code review, a first look at memories, one-click MCP setup, Jupyter support, and general availability of Background Agent. Automatic code review with

You can now launch Background Agents directly from Slack by mentioning **@Cursor**. Agents can read the thread, understand what's going on, and create PRs in GitHub, all without leaving the conversati

Agent To-dos Agents now plan ahead with structured to-do lists, making long-horizon tasks easier to understand and track. The agent breaks down longer tasks with dependencies, visible to you in chat a