Yogin Patel
791 posts
Yogin Patel
@Yogin16
AI @sprinklr. Thoughts my own!
Dubai Katılım Ocak 2012
1.1K Takip Edilen257 Takipçiler
@zebassembly and take that too at extreme
Yogin Patel@Yogin16
loop that can trigger other similar loops but also fix parent loop but all running without any hierarchy as their self-contained loops yet somewhat aware of the protocols of the loop Loop looping and creating loops for the Loop to be looping.
English
verifiers
renderers
tasksets
harnesses
codebases
worldsims
…
open superintelligence stack
English
“The person is not a system of algorithms.”
— Pope Leo XIV
English
weird from cc: everytime cc tries to integrate llm call, raw or from frameworks (tried it on agno as well as langchain), it would try its best to hardcode claude! even though framework supports model agnostic and you want to continue to support them! observed thrice!
English
opencode go is pretty solid setup for start
but context length of the model could be a problemo
Yogin Patel@Yogin16
where are you all hosting your kimi k2s? after claude and codex both maxed out on limits looks like need local setup?
English
where are you all hosting your kimi k2s?
after claude and codex both maxed out on limits looks like need local setup?
Kimi.ai@Kimi_Moonshot
Meet Kimi K2.6: Advancing Open-Source Coding 🔹Open-source SOTA on HLE w/ tools (54.0), SWE-Bench Pro (58.6), SWE-bench Multilingual (76.7), BrowseComp (83.2), Toolathlon (50.0), Charxiv w/ python(86.7), Math Vision w/ python (93.2) What's new: 🔹Long-horizon coding - 4,000+ tool calls, over 12 hours of continuous execution, with generalization across languages (Rust, Go, Python) and tasks (frontend, devops, perf optimization). 🔹Motion-rich frontend - Videos in hero sections, WebGL shaders, GSAP + Framer Motion, Three.js 3D. 🔹Agent Swarms, elevated - 300 parallel sub-agents × 4,000 steps per run (up from K2.5's 100 / 1,500). One prompt, 100+ files. 🔹Proactive Agents - K2.6 model powers OpenClaw, Hermes Agent, etc for 24/7 autonomous ops. 🔹Claw Groups (research preview) - bring your own agents, command your friends', bots & humans in the loop. - K2.6 is now live on kimi.com in chat mode and agent mode. For production-grade coding, pair K2.6 with Kimi Code: kimi.com/code - 🔗 API: platform.moonshot.ai 🔗 Tech blog: kimi.com/blog/kimi-k2-6 🔗 Weights & code: huggingface.co/moonshotai/Kim…
English
okay switched to pi
i think it is /tree and the token efficiency that is making people talk about it
still can imagine cc for plan mode and codex harness for adversarial review
English
Today we're announcing our Series C funding: $355M at a $4.65B valuation, led by some great investors @generalcatalyst and @Redpoint.
We've had insane growth in the last year, but we're still very early. So proud of the team and what we have built so far!
Modal@modal
English
@thayto_dev @superset_sh Superset definately better than conductor
But recent updates have been choppy
Trying opencode also since today
English
I've been advocating @superset_sh to all my friends and coworkers, and they prefer it over Conductor
imagine you have to stop all of your claude/codex sessions just to run an update because your tool (conductor) can’t run in the background
English
Yogin Patel retweetledi
it’s in gemini, just create it in ai studio. oh, that’s for your personal google one account. for workspace you need gemini business. no, not gemini advanced, that’s ai pro now. unless you need ai ultra. oh agents? you do that in spark actually. no, not gemini api managed agents, that’s different. for coding use jules. unless you mean the agentic ide, that’s antigravity. no, that’s the old antigravity, download the new one. actually gemini cli is being deprecated, use antigravity cli. no the flash model is smarter than the pro model. unless you need pro. if it’s video, use flow. no, flow uses veo. no, nano banana is images. actually that’s in gemini now. unless you’re in search, then it’s ai mode. no, research is notebooklm. anyway it’s all very simple.
English
next trillion dollar company is someone teaching models equivalent of sleep
English
the reason ai adoption in software is higher than rest is not that engineers love to build stuff for themselves
it is because software was already consolidated so much that context gathering was reduced to a simple grep
the great consolidation is prerequisite for great automation
English
How to understand the x algorithm in 60 seconds:
1. `git clone github.com/xai-org/x-algo…`
2. `codex /goal thoroughly understand the drivers behind the x
algorithm and deliver a concise set of recommendations for growing on the platform`
Elon Musk@elonmusk
The latest 𝕏 algorithm has been published to GitHub github.com/xai-org/x-algo…
English