Tried this on Dots with Astra x Blender to compare against Opus. I used the same quote that Leon used then added some animals. Opus did quite well with the animals
Opus 5.5 built a tiny world in a glass box
you can: grab a cloud > squeeze rain out of it
shake it > lightning
"draw" wind using the wind brush
and theres a day night cycle + seasons as well
claude.ai/artifact/3bhygX763…
1
34
A fine feature. Works great as cheap parsers. Fuzzy, intent, cascading.
here's how quickly we move at @OpenAI:
Sept 22 - initial Decisions API prototype
Sept 29 - DevDay announcement
today - launch 🚀
just two weeks from local demo to public launch
8
Deon retweeted
here's how quickly we move at @OpenAI:
Sept 22 - initial Decisions API prototype
Sept 29 - DevDay announcement
today - launch 🚀
just two weeks from local demo to public launch
the general trend for tech influencers nowadays are to quickly praise new solutions on day 1 and then shortly take back their praise. Not a great direction. Always been skeptical of evaluators who have dealt with the models in volume.
i take back what i said about opus 5
initial results were promising / divergent but the more time i spent with it, the more infuriating of an experience it became
my team feels the same way
i am back on gpt-5.6-sol as my daily driver with fable and kimi 3 on planning / reviews
20
Deon retweeted
Some people seem to be surprised that an old geezer like me could adapt so quickly to the newest tech.
The thing is, I've been a programmer for over 60 years. And every single year brings another adaptation to new tech. Adapting to new tech is what we programmers do on a regular basis. Why would I stop now?
Deon retweeted
Notably, the budget panel was comparable with Claude Fable 5 in performance.
A panel of Gemini 3 Flash, Kimi K2.6, and DeepSeek V4 Pro, fused together, beat solo GPT-5.5 and solo Opus 4.8 outright.
And it landed within 1% of Fable 5 while costing roughly half the price.
Okay 3 cent for flash... this was what Google was supposed to charge for Gemini-3.1-flash-lite
Replying to @deepseek_ai
API is Available Today!
🔹 Keep base_url, just update model to deepseek-v4-pro or deepseek-v4-flash.
🔹 Supports OpenAI ChatCompletions & Anthropic APIs.
🔹 Both models support 1M context & dual modes (Thinking / Non-Thinking): api-docs.deepseek.com/guides…
⚠️ Note: deepseek-chat & deepseek-reasoner will be fully retired and inaccessible after Jul 24th, 2026, 15:59 (UTC Time). (Currently routing to deepseek-v4-flash non-thinking/thinking).
6/n
92
Deon retweeted
API is Available Today!
🔹 Keep base_url, just update model to deepseek-v4-pro or deepseek-v4-flash.
🔹 Supports OpenAI ChatCompletions & Anthropic APIs.
🔹 Both models support 1M context & dual modes (Thinking / Non-Thinking): api-docs.deepseek.com/guides…
⚠️ Note: deepseek-chat & deepseek-reasoner will be fully retired and inaccessible after Jul 24th, 2026, 15:59 (UTC Time). (Currently routing to deepseek-v4-flash non-thinking/thinking).
6/n
Yes anyone can make a MVP app. Now what about the other 20 steps afterwards. I still spend so much time architecting the right production design and requirements. Would be easy if I had unlimited money in AWS. Wouldn’t that be nice. Building against each enterprise constraints is what separates hobbyists and builders.
20












