Nurok AI

The living knowledge layer of Agentic AI

[Nurok AI @nurokhq]

https://t.co/DOCEkJPbob

[Kimi.ai @Kimi_Moonshot]

Releasing the model weights and technical report of Kimi K3.

Kimi K3 is our most capable model: a 2.8T MoE model with native visual understanding and a 1M-token context window.

New model architecture: 2.5x the intelligence per unit of compute, not just more params.

Alongside Kimi K3, we're opening up more of the stack behind it — high-performance attention kernels, MoE communication library, and infrastructure for running agent environments at scale.

Model weights: https://t.co/7m7eEg6Y0B
Tech report: https://t.co/yeu6cjpMCT
Tech blog: https://t.co/YTfiMSNM1f

[Nurok AI @nurokhq]

There's a name for what AI agents actually need.
Not a RAG pipeline.
Not a vector store.
Not a Git repo full of markdowns.

An Agentic Knowledge Base (AKB).
A distilled, sourced corpus that stays current and is portable to any agent in any tool.
Because every claim is sourced, you can finally trust the agent to handle a workflow instead of just hallucinating facts you can’t verify.

We're fixing the loop.
Keep an eye on it.
@nurokhq

#AKB #AgenticKnowledgeBase #Nurok #AIAgent
#MCP #KnowledgeManagement #AI

[Nurok AI @nurokhq]

https://t.co/wfwm8xuVlN

[Nurok AI @nurokhq]

Re-researching the same topic on every query is expensive.
Distill once.
Consume many times.
Published comparisons put stored retrieval at 8 – 80× cheaper than on-demand research.
And sub-second vs tens of seconds.
The economics of maintained knowledge are obvious.
The infrastructure to make it work isn't built yet.

#AgenticAI #AI #Distill

[Bhavy☄️ @Bhavani_00007]

I tested Kimi K3 vs Claude Opus 4.8

Same prompt, an armory bay with lighting, props, and detail. Top is Kimi K3, bottom is Opus 4.8.

It's not even close.

Kimi K3 built a full scene with textures, proper lighting, ammo crates, weapon racks, working detail everywhere. Opus 4.8 gave me a near empty room with a couple of floating tables.

No doubt it beats Opus 4.8. Kimi K3 is Fable 5 level, and it's clearly better than GPT-5.6 Sol at 3D and games.

An open weight model just matched the best closed models on the market. Let that sink in.

[Nurok AI @nurokhq]

https://t.co/oGY3lN7qis

[Nurok AI @nurokhq]

Every model has a training cutoff.
The moment it ships, it starts going stale.
Most people patch this with one-off research:
search, read, paste into context, repeat next week.
We're building the layer that stays current automatically. Follow along.

#Nurok #ComingSoon

[Nurok AI @nurokhq]

Building a knowledge base for your agent today means stitching together crawlers, vector stores, Git repos, and cron jobs by hand.
It leaks at every step.
And it goes stale the moment you save it.
Nurok closes the loop.

[Nurok AI @nurokhq]

60% of consumers already click AI-generated results when searching for local businesses.
That number is growing every month.
Most businesses have no structured presence for AI to find.
That's the gap.

[Nurok AI @nurokhq]

Google knows more about local businesses than anyone.
But that data isn't built for AI Assistants.
It's built to keep you on Google.
No contact layer. No MCP. No deep enrichment.
That's the gap Nurok is built for.