From 0s & 1s to infinite possibilities.
- Tweets 7,066
- Following 1,779
- Followers 1,980
- Likes 2,026
Pinned Tweet
Finally its here...get the complete vibecoding starter kit with templates and prompts! enjoy - repo in the comments
8
6
85
11,075
OpenAI is offering ChatGPT plus free for most of the old accounts who never took a paid plan… so if you have ever signed up using an email ID … try logging in to see the offer!
2
87
Opencode desktop is cool but I see a problem….
When sub agents are initiated you don’t know how many sub agents are working unless you manually hover around the icon… it will be great if sub agents visibility can be added on screen by default?
@thdxr
58
Imagine living in this area where highest income tax payers are living! Probably roads in Iran look better than this… what a time to be alive!
Gunjur-Varthur Road: Work is progressing and change is visible.
Essential tasks including waste clearance, pothole repairs, the removal of electric poles and roadside obstructions, and road levelling have been accelerated.
1
49
Antigravity adds Opus 5.5 and Sonnet 5.5 but only for paid Pro and Ultra users!
So if you have Pro sub given to you for which you didn’t pay ( example- via Jio in India), you can’t see these models.
Also the current 3rd party models will disappear on 2nd November!
Bad move @_mohansolo
1
1
192
GPT 6.1 Sol was slow so I tried GPT Astra… my limit was down from 15% to 0% just with context compaction!
What the hell….
1
124
Deepseek desktop harness is here!
Please check out @DeepSeekHarness. We just released packaged desktop versions for macOS and Windows; Linux users can get it from the @deepseek-ai/dsh package on npm. deepseek.com/harness
83
Wow! Human looking avatar in video that too real time!!!!
Introducing Griffin, the first model to pass the video Turing test.
48% of people who talked to it live thought it was a real human. Previous systems have had a pass rate <3%. It is #1 on NVIDIA's benchmark for full-duplex AI video.
It’s the first Human Interaction Model (HIM).
Community note
The 48% figure and "video Turing test" claim are from Tavus's own study of 54 one-minute calls, not independently verified or using a standard protocol. Griffin-Lite leads NVIDIA's VideoFDB benchmark on their public leaderboard.
cellcog.ai/blog/tavus-gri…
research.nvidia.com/labs/amri/proj…
tech-ish.com/2026/10/02/tav…
88