Thilak @webtkdev

I was a big fan of Grok models. 4.7 really messed it up. I’m not saying it is poor; it is good for writing or reading, sometimes good in a coding workflow, but it can’t be trusted all the time at the moment. Better release 4.8 soon.

Opus 5.5 > Sol 6 > Grok 4.7

Here we go: Sonnet 5.5 already being stealth tested. The model has a 1M-token context window and a 128K-token maximum output. Pricing per 1M tokens: $2 input, $10 output. This reads to me as a direct response to GPT-6-Sol. Same pricing and presumably similar performance (if not better). The battle continues!

lyra @lyraxana

a day ago

Anthropic is currently stealth testing Claude Sonnet 5.5 (`claude-sonnet-5-5`). Check comments for more information meow

35 23 1K 296K 93

Opus 5.5 is so much better than Sol 6. It offers much better usage. OpenAI this time didn't compete. Reducing the cost is okay, but performance is not comparable to Opus 5.5. They reduced the price as well. But it is just amazing, one-shotting anything I throw at it.

In today's release, Opus 5.5 wins in the performance and GPT 6 Sol & Luna wins in the cost.

Browser Use Bench v2 Pareto frontier got completely redrawn today > Claude Opus 5.5: 59.4 > GPT‑6 Sol medium: 66.9 (3.5x cheaper) > GPT‑6 Luna xhigh: 57.6 (22x cheaper than Opus) OpenAI is in its own league 🔥 All models available to try on our cloud. [x-axis is log scale]

So who is the today's winner? GPT 6 Sol vs Opus 5.5

GPT 6 Sol and Luna arrived with 50% cost cut. openai.com/index/introduc…

Sonnet 5.5 and Haiku 5.5 is coming soon. 🔥

Best ever release from Anthropic and it feels more friendly.

Claude @claudeai

3 days ago

Introducing Claude Opus 5.5, the first model in our new Claude 5.5 family. It performs at the level of Claude Fable 5.1 for most tasks, and costs 40% less to run than Opus 5.

3K 9K 95K 24.8M 13K

❤️❤️❤️

Google @Google

4 days ago

Introducing Googlebook, a new category of laptop, available for preorder today. 💪 Crafted with 2.8K OLED touchscreen displays, 14 hour battery life and all the performance needed to power your ideas 📳 Engineered to sync effortlessly with your @Android phone, so you can jump

515 587 7K 1.2M 1K

Grok 4.7 is out.

BREAKING CMF is becoming an Indian brand!! Carl Pei just announced that CMF is spinning out of NOTHING. - Majority Indian-owned - HQ + R&D in India - NOTHING stays as a shareholder I hope they bring a better product lineup next year..

Meet Qwen3.8-LiveTranslate, Qwen's next-generation real-time simultaneous interpretation model! 📢 Built on an Interleave architecture, it improves faithfulness, fluency, and conciseness while reducing average lagging (LAAL) from 2.8s to 2.3s across 60 languages. New capabilities: 🙌 - Real-time speaker diarization — distinguishes speakers in multi-party speech and preserves each speaker's voice through more stable voice cloning. - Synchronized bilingual display — source and translation on screen together. - Long-context disambiguation — leverages conversation history to clarify names and terminology for consistent translations. Let's try Qwen3.8-LiveTranslate! 🥳 - Blog: qwen.ai/blog?id=qwen3.… - QwenCloud: qwencloud.com/models/qwen3.8…

Meet Qwen-Image-2.1, the most balanced and cost-effective image generation model in the Qwen-Image series! Now open weights! 🎨 A unified model for both generation and editing, delivering top-tier quality in a lightweight package. Highlights: 👀 - Compact & exceptionally fast: A lightweight 7B architecture that outperforms most closed-source models, with drastically accelerated inference for multi-image inputs. - Native transparency: Natively generates and edits RGBA layers, enabling seamless compositing and text editing within transparent images. - Versatile, high-fidelity editing: Supports up to 10 reference images and precise local control while preserving strict fidelity for portraits and products. - Broad coverage & stunning aesthetics: Excels at panoramas, infographics, and virtual try-ons, delivering realistic textures and elegant typography. Start to create your next masterpiece with Qwen-Image-2.1! 🖼️ - Blog: qwen.ai/blog?id=qwen-i… - GitHub: github.com/QwenLM/Qwen-Im… - Model Scope: modelscope.cn/models/Qwen/Qw… - Hugging Face: huggingface.co/Qwen/Qwen-Imag…

We're adding support for AGENTS.md to Claude Code. Starting today in version 2.1.277, if there is no CLAUDE.md in a folder, Claude will check for and use AGENTS.md. You can toggle this behavior in /config.

Few days back Cursor launched the same. Now Claude. Probably OpenAI will launch the same next week.

Claude @claudeai

a week ago

Projects now run from one conversation, starting in Claude Code. You describe what needs doing, and Claude directs parallel threads that keep working after you close your laptop. In beta today for select Pro and Max users in cloud sessions; coming to all Claude users soon.

597 766 12K 3.5M 5K

@chrisreedbates Good suggestion. Like pets, we can have this.

Codex usage on the $20 Plus plan feels seriously nerfed lately. Over the last few days, I’ve noticed my 5-hour usage limit getting exhausted ridiculously fast, even when using Luna or Terra. I’m not sure if OpenAI reduced the available usage/bonus credits behind the scenes or if the 5-hour window itself is the problem, but something definitely feels different compared to before. For a while, when usage was primarily governed by the weekly limit, Codex felt much more usable. With the 5-hour limit back in the picture, I’m burning through the available usage so quickly that Plus is starting to feel more like a plan for using ChatGPT as an assistant rather than doing serious coding work with Codex. At this point, for actual day-to-day coding, Cursor feels like much better value to me. No annoying 5-hour window, and I can actually keep working without constantly watching a usage meter. It almost feels like serious Codex users are being pushed toward the $200 plan. And surprisingly, comparing the ~$20 plans, even Claude’s 5-hour usage window feels more usable to me right now. Something has definitely changed with the Codex Plus experience, and not in a good way.

@gregpr07 Omg! This is insanely fast. Gonna try this asap.