Dipankar🧑‍💻

Software Engineer. Tech @Deel

[Dipankar🧑‍💻 @dipankarcodes]

DeepSeek is ridiculously cheap.

GPT-5.6 → $2.47
DeepSeek → $0.05

How DeepSeek pulls off this pricing is wild. 50x cheaper than GPT 5.6 with similar output.

[Dipankar🧑‍💻 @dipankarcodes]

Free users get unlimited GPT-5.6 Luna starting tomorrow.

Anthropic when ?

[OpenAI @OpenAI]

We’re making better intelligence easier to access in ChatGPT for everyone:

- GPT-5.6 Sol now powers both Instant and deep reasoning for Plus & Pro users, delivering more factual, focused responses.

- Free & Go users get unlimited text chats with GPT-5.6 Luna starting tomorrow.

[Dipankar🧑‍💻 @dipankarcodes]

@Da7_Tech Lol no. It’s better than Luna, but not Sol

[Dipankar🧑‍💻 @dipankarcodes]

Almost the entire growth story is concentrated in one corner.

Concerning.

[Eric Basmajian @EPBResearch]

92% of the economy is growing 1%

8% of the economy is growing 14%.

[Dipankar🧑‍💻 @dipankarcodes]

Meta’s Muse Spark 1.2 just beat Opus 5 on a hard finance agent benchmark while being 6.7× cheaper and nearly twice as fast.

Zuck is back at the AI race

[Vals AI @ValsAI]

Muse Spark 1.2 is the first model to crack 60% on Finance Agent v2, our benchmark that gives models the job of a financial analyst.

At $0.77/test it is 6.7x cheaper than the previous #1, Opus 5 ($5.12), at twice the speed.

[Dipankar🧑‍💻 @dipankarcodes]

@XFreeze Grok 4.5 is pretty good at 3D as well
https://t.co/1SRnlhly98

[Dipankar🧑‍💻 @dipankarcodes]

Grok 4.5 is absurdly good at 3D.

Total cost to build this 3D lesson: $0.80

It’s getting crazy how fast you can now ship real technical products for almost nothing - and the costs keep getting cheaper.

[Dipankar🧑‍💻 @dipankarcodes]

Grok 4.5 is absurdly good at 3D.

Total cost to build this 3D lesson: $0.80

It’s getting crazy how fast you can now ship real technical products for almost nothing - and the costs keep getting cheaper.

[Dipankar🧑‍💻 @dipankarcodes]

@ValsAI Meta starting the price war now

https://t.co/Gsr7JYm3oV

[Dipankar🧑‍💻 @dipankarcodes]

You can run Muse Spark 1.2 for almost nothing if you let them train on your data.

Contributor version: $0.10 input / $0.20 output
Normal: $1.25 / $4.25

Same 1M context.

12-20x cheaper for the same model. Absolute steal.

[Dipankar🧑‍💻 @dipankarcodes]

@CommandCodeAI Kimi 3 is the winner. But pricing is pretty good for Muse Spark

https://t.co/Gsr7JYm3oV

[Dipankar🧑‍💻 @dipankarcodes]

You can run Muse Spark 1.2 for almost nothing if you let them train on your data.

Contributor version: $0.10 input / $0.20 output
Normal: $1.25 / $4.25

Same 1M context.

12-20x cheaper for the same model. Absolute steal.

[Dipankar🧑‍💻 @dipankarcodes]

@MetaforDevs Pricing is insane here
https://t.co/Gsr7JYm3oV

[Dipankar🧑‍💻 @dipankarcodes]

You can run Muse Spark 1.2 for almost nothing if you let them train on your data.

Contributor version: $0.10 input / $0.20 output
Normal: $1.25 / $4.25

Same 1M context.

12-20x cheaper for the same model. Absolute steal.

[Dipankar🧑‍💻 @dipankarcodes]

@cryptopunk7213 Pricing is also interesting

https://t.co/Gsr7JYm3oV

[Dipankar🧑‍💻 @dipankarcodes]

You can run Muse Spark 1.2 for almost nothing if you let them train on your data.

Contributor version: $0.10 input / $0.20 output
Normal: $1.25 / $4.25

Same 1M context.

12-20x cheaper for the same model. Absolute steal.

[Dipankar🧑‍💻 @dipankarcodes]

You can run Muse Spark 1.2 for almost nothing if you let them train on your data.

Contributor version: $0.10 input / $0.20 output
Normal: $1.25 / $4.25

Same 1M context.

12-20x cheaper for the same model. Absolute steal.

[Mark Zuckerberg @finkd]

Releasing Muse Code in beta today. It's a terminal coding agent that takes on complete software engineering tasks across large repos: planning changes, writing code, validating the results. Powered by Muse Spark 1.2, a coding-focused model update.

[Dipankar🧑‍💻 @dipankarcodes]

GPT 5.6 Luna sucks at coding.

If you actually want cheap models with real performance per dollar, just run DeepSeek Flash V4.

[Dipankar🧑‍💻 @dipankarcodes]

Testing is still the bottleneck

[Dipankar🧑‍💻 @dipankarcodes]

The MapReduce / TensorFlow / Gemini core just left Google.

Jeff Dean, Sanjay Ghemawat, Oriol Vinyals, and Quoc Le are founding Discovery Loop - a public benefit corp to fully automate the experimental loop in ML, science, and engineering.

It's over for Google

[Jeff Dean @JeffDean]

Announcing Discovery Loop!

I am very excited to announce that, along with my longtime friends and collaborators @Sanjay_Ghemawat, @OriolVinyalsML and @quocleix, we are founding Discovery Loop (@DiscoLoopAI), a Public Benefit Corporation whose mission is to automate machine learning, science, and engineering to accelerate discoveries and progress. The four of us have worked together for 14 to 30 years, and have helped build some of the world’s most used products, infrastructure and AI models, and we’re excited to turn our attention to this ambitious endeavor.

♾

Learn more at: https://t.co/Rv3LMdLluK

[Dipankar🧑‍💻 @dipankarcodes]

Another banger from brad pitt

[Prayag @theprayagtiwari]

Bolte bolte 'Maun Vrat', Lord Ravi kishan for a reason 🔥

[Alex Bouaziz @Bouazizalex]

Excited to welcome @getclarity_ai to Deel!

AI has changed how companies need to approach security and combat fraud. That’s why we’re bringing the incredible Clarity team in-house to build one continuous identity security platform, from recruitment to onboarding to management.

https://t.co/E008dZTDaL

[OpenAI Newsroom @OpenAINewsroom]

Apple is getting this wrong.
https://t.co/IStp6WhOrS

[Dipankar🧑‍💻 @dipankarcodes]

DeepSeek so cheap it almost didn’t make the chart.

[Andrew Curran @AndrewCurran_]

At first I thought Bloomberg forgot to add DeepSeek's pricing to the chart.

[Dipankar🧑‍💻 @dipankarcodes]

Cool but when’s it dropping on Zen ? @opencode

[OpenCode @opencode]

Qwen3.8 Max is now available in OpenCode Go

[Dipankar🧑‍💻 @dipankarcodes]

Deel just crossed $1.5 billion ARR 📉💰

From zero to one of the fastest-growing global workforce platforms on the planet. And we’re still accelerating - AI, identity security, and the next chapter already underway.

Proud to be a part of this rocketship 🚀