
Building in AI x Finance | CSE grad @IITKGP | prev @witnesschain
Joined December 2018
- Tweets800
- Following969
- Followers300
- Likes1.7K
Just discovered that MCPs/agent APIs serve as a neat backdoor to some open-source licenses. Take AGPL, for example: you’d typically have to share source code if you distribute it. But if you sell an agent using this software via MCP (unmodified source software kept internal) …✨
1
45
Clear shift towards open models in the past 1 year, the % share is widening in their favour
While not all traffic is accounted for by just looking at open router, it’s definitely indicative of the shift. Big labs have to continue pushing for cost efficiency!
1
77
Been migrating to open models, since pricing of claude models on API is just not sustainable.
GLM was meh, deepseek v4.1 flash is the current choice. What’s the optimal setup for it ? - high TPS with reasonable cost
Rent GPUs, Buy GPUs, some provider API?
3
1
126
Well well well ….
Nice DeepSWE score! Does it survive a lowercase filename?
Introducing MaxxBench: 89 DeepSWE tasks with small changes to the instructions or starting code. Same engineering problem, slightly different requirement.
Models clustered around 74% on DeepSWE end up 23 pp apart.
Good at coding, or good at DeepSWE? 🧵
55
Been waiting for IDEs to do build in this direction , excited to see first one’s already here!
Voice mode in Antigravity is here. Enabled by the Gemini Live model, you can use your voice to plan, steer code changes, and approve permissions out loud. Have a live conversation across multiple languages, interrupt whenever you need to, set your preferred voice and playback speed, and when you're done, just say "bye" or tap end. Happy building!
47
Google is bacc? 🔥
I did say we'd be releasing our models periodically...introducing Gemini 4 Carbon, our next frontier model.
Following Argon's release just two weeks ago, Carbon shows the pace at which we’re pushing the frontier, it ranks #1 on the recently updated v5.0 of @ArtificialAnlys Intelligence Index at 59. It also leads across key cybersecurity evaluations scoring 88.2% on SEC-bench Pro for finding vulnerabilities, 54.8% on ExploitGym for exploitation, and 71.0% on CWE-bench for patching.
We’re seeing significant improvements in long-horizon tasks even compared to Argon, and we’ve strengthened frontier safeguards.
As we did with Argon, we’re going to roll it out gradually + responsibly. Some trusted cyber defenders will get access starting this week. We'll be expanding it to more partners through our Fairwind Program, ahead of wider availability soon, so hang tight!
1
1
65
They don't compare with the SoTA anymore - astra, opus 5.5 etc.
Saw similar benchmark release pattern from DeepSeek.
Has china fallen behind again or just flood cheap flash models (Cost Frontier), and saving the Intelligence Frontier launch for year end finale? 👀
MiniMax M3.1 Flash: Space Performance + Bunny Speed
🚀🐰 Yes, Space Bunny - which has topped global leaderboards with 200T+ tokens - was an earlier version of M3.1 Flash.
💰Redefining the performance economics of agentic AI. M3.1 Flash achieves frontier-level intelligence across key tasks such as Cowork (Automation-Bench: 62%; ALE: 30%), Coding (DeepSWE: 74%) and Visual Reasoning (BabyVision: 91%), at 1/3 the cost of M3 and 3x the speed.
API pricing (flat rate): $0.15 / $0.60 per 1M input/output tokens, with cache reads at $0.015 / 1M tokens.
Model weights will arrive on Oct 23.
API: platform.minimax.io
M Plan: minimax.io/m-plan
MiniMax Code for desktop: agent.minimax.io/download
50
Pankaj Agarwal retweeted
#India: We are concerned by reports of mass detentions of demonstrators, protest leaders, members of civil society organisations, lawyers and journalists, before and during today’s large-scale demonstrations in Delhi.
We call on the authorities to ensure that the right to peaceful assembly is fully respected and fulfilled.
ohchr.org/en/press-releases/…
Yep, serious bot detection will be required soon as these agents are commodotized. Muse is first step towards a potential disaster if not managed well!
Hello computer,
Find every Amazon purchase I ever made.
Call the number of the manufacturer.
Say you’re not satisfied.
Request a refund.
If they decline, threaten to leave a bad review.
If they decline, find the CEO’s phone number and ask him for a refund.
Now, repeat the same process for items I never even purchased and see if you can get them to send me money.
If they ask for a recipient, generate an image of one and send that.
Do not stop until you’ve collected $1 million in refunds.
78
Opus 5.5 is impressive, pretty much fixes all the problems I had with opus 5.
No overthinking, no gibberish long unreadable text mountains, much more apt, faster, and indeed cheaper on usage limits.
Fable 5.1 is literally useless now, I expect they push a new Fable soon too!
61
Finally, it might be good again, gonna go check this 🤠
Replying to @claudeai
Opus 5.5 communicates more naturally, addressing some of the most common feedback we heard on Opus 5.
It puts the most important information up front and follows the writing rules you give it, which makes long sessions easier to follow.
70
Just me or usage limits draining faster on @claudeai today?
Applies to both fable & the common pool - it's been significantly worse for me today compared to the previous week
1
2
62
Terminal 2.1 vs terminal 4 diff, especially given the relative consistency of opus 5, is kinda hinting at benchmaxxing 🙃
Even if not, at the very least, it's not optimised for the right metrics which make it more generalizable than benchmarks
Today we’re introducing a new 3.8 Flash model, our 3rd Flash release in just 6 wks.
It delivers significant leaps from 3.7 Flash across software engineering, agentic tasks, and multi-step reasoning. On DeepSWE v1.1, it outperforms most larger frontier models in autonomously solving complex engineering problems end to end, at a fraction of the cost.
More details here: blog.google/innovation-and-a…
1
86
Wow!
🚨JUST IN: China surprises the world by announcing a cure for Type 1 and Type 2 Diabetes using Stem Cells and allowing the body to produce Insulin again
Readers added context they thought people might want to know
Chinese researchers have achieved promising results using stem cells to help some patients with Type 1 and Type 2 diabetes produce their own insulin. These experimental treatments are in clinical trials. China has not announced a cure for diabetes.
scmp.com/news/china/sci… chinacarehealthtours.com/stem-cell-ther… verifica.efe.com/china-no-ha-an…
36
Pankaj Agarwal retweeted
Qwen3.8 is launching and going open-weight soon!🌐
With a massive 2.4T parameters, this model is continuously evolving. We believe it’s one of the most powerful model available today, compatible to leading frontier AI models , second only to Fable 5.
You don't have to wait to test it. Just now, the Qwen3.8-Max-Preview made its debut on Alibaba’s Token Plan, Qoder, and QoderWork. Be among the very first to try it out.
Can't wait to hear what you build. Stay tuned! 🚀
Token Plan
international:qwencloud.com/pricing/token-…
China:platform.qianwenai.com/prici…
Big news: Kimi-K3 by @Kimi_Moonshot is now #1 in the Frontend Code Arena with 1679 pts, surpassing Claude Fable 5.
This is a 17-place jump from Kimi-k2.6 (#18 -> #1).
In Frontend, Kimi-K3 ranked #1 in 6 of 7 domains: Brand & Marketing, Reference-Based Design, Data & Analytics, Consumer Product, Simulations, and Content Creation Tools, landing #2 only in Gaming behind Fable 5.
The full model weights will be released by July 27.
Congrats to the @Kimi_Moonshot team on this major milestone!



























