Shiva

Can't keep up with AI Updates | AI Shit Posting

[Shiva @LShivkumarreddy]

Google's Official AI Agents Course.

It's a 5 day intensive training to know from 0-1 of building real-world ai agents that works and scale.

Link: https://t.co/iSIqf7B912

[Google Cloud Tech @GoogleCloudTech]

Level up your AI skills with the 5-day generative AI intensive course by Google Cloud and @kaggle.

Join over 1 million learners for this no-cost online course starting today! Sign up now to build production-ready agents and secure your capstone prize → https://t.co/c1t0oyNxcH

[Shiva @LShivkumarreddy]

That explains why I am always out of my tokens & limits!

[Composio @composio]

Hermes and Pi Agent led on the average cost per task, while Claude Code cost about 3.7x as much as Pi:

- $0.39 Hermes Agent
- $0.40 Pi Agent
- $0.47 Codex
- $0.51 OpenCode
- $0.54 Kimi Code
- $1.47 Claude Code

The median cost tells the same story: $0.29 in Pi Agent and Hermes, $0.35 in OpenCode, $0.38 in Kimi Code, $0.39 in Codex and $0.72 in Claude Code, so the cost gap holds for a typical task and is not driven by a few expensive runs.

We calculated these costs using Kimi K3’s list prices: $3/1M input tokens, $0.30/1M cached input tokens, and $15/1M output tokens.

[Shiva @LShivkumarreddy]

@ghagler @rileybrown @jack It’s true! I guess it’s valid when you use block or buzz hosted relays. If you host your own relay I think all the data collection will be mitigated.

Of course you cannot be public!

[Shiva @LShivkumarreddy]

@hot_town I kinda get what you are trying to do!

But I am excited for this!✨🔥

[Tom Brow @_tombrow]

The iOS and Android apps are actual Buzz, not a remote control that needs your laptop open

First half is my practical take on how Buzz is different from other chat with agents. Without saying "nostr" or "sovereign" even once

https://t.co/YLKmKIRBWt

[Shiva @LShivkumarreddy]

@deedydas I think I saw something similar somewhere today!

[Vinny @hot_town]

I wrote up an agent-friendly guide so you can get the exact same workflow as in the video below.

Just copy this raw text and give it to your agent:

https://t.co/dGURqp8udx

yer welcome.

[Vinny @hot_town]

I made a presentation on Nostr for this video but only ended up using 3 slides.

It’s called “Nostr for Fullstack Devs”.

Here’s the link to it (23 slides total) if you wanna learn about the protocol behind Buzz and how to build on it:

https://t.co/1t2MJAHVxc

[ThomPete @Hello_World]

For those who want to understand how Buzz for Mobile works a little better here is @_tombrow with a deeper dive.

Most important piece of information. Buzz Mobile is actually Buzz, not just a remote that needs your laptop open.

https://t.co/NMM3IaPpmb

[Kevin Simback 🍷 @KSimback]

The most under-discussed feature of Buzz is Projects - it's basically an integrated GitHub right into Buzz

No more creating GitHub tokens and managing repo permissions for your agents working with code

Clearly not a robust replacement for GitHub but great for small projects

[Pavlenex @pavlenex]

Announcing BuzzDir, a directory of publicly shared Buzz communities.

Buzz is best experienced in multiplayer, but finding your hive can be a challenge.

🐝 FOSS, built entirely in Buzz
🍯 Discover or list communities
🛠️ Improve BuzzDir through Buzz

https://t.co/OJ8fZfvfaw

[Shiva @LShivkumarreddy]

@wesbillman This is really awesome! Right on it with my templates too!

[Shiva @LShivkumarreddy]

All the buzz lately about @jack’s Buzz got you to try it with @NousResearch’s Hermes harness and you got hooked!

But you have been building your own personal agent with hermes and wanted to use the same?

Try the below skill!

[mr-r0b0t @mr_r0b0t]

Many of you asked for @NousResearch Hermes Agent profiles that would be durable across both Hermes and Buzz.

Good news, there's a skill for that!

Make a profile in Hermes, push it directly to @blocks Buzz, access it in either location!

https://t.co/WysfEeHs1I

[Shiva @LShivkumarreddy]

This is surreal!

I think this has been a recent between two parties that don’t wanna point out in national television! lol🤣

When rolling out to GA @SarvamAI?

[Sarvam @SarvamAI]

Introducing Saaras V4 Multi-speaker

Our best-in-class speech recognition model that accurately captures overlapping conversations and multi-speaker interactions, while delivering state-of-the-art performance in English across accents.

[Shiva @LShivkumarreddy]

Please tell me someone’s reading the script from behind. @SarvamAI nailed it!

[Sarvam @SarvamAI]

Presenting Bulbul V4

Our Text-to-speech model with richer emotion, natural expression and greater vocal range.

Give it a listen:

[Shiva @LShivkumarreddy]

@soul4tech @WhaleInsider +1

[Shiva @LShivkumarreddy]

I think @SarvamAI is renting from phone numbers from @Vobiz_Ai

Or is it just KYC not sure!

[Rajan Anandan @RajanAnandan]

Up and running in 30 secs! Fully self serve!

[Shiva @LShivkumarreddy]

@SarvamAI didn’t promise anything and delivered a lot of value today.

Beginning to see a lot of value for the org!

Keep it up!

[The Offline Network @OfflineOnAir]

Everything Sarvam announced at Epoch:

MODELS
1/ Sarvam 105B (upgraded)
2/ Sarvam Vision 2.0: doc AI with native key-value extraction
3/ Saras V4 (speech-to-text): now SOTA on global English
4/ Saras V4 multi-speaker: transcribes people talking over each other
5/ Bulbul V4 (text-to-speech): one of India's most expressive TTS

INFRA
6/ Sarvam Inference: serves top open models from Indian soil
7/ Custom model training: your own model on Sarvam's infra
8/ Compute: now running the largest Blackwell cluster in India
9/ Opening a San Francisco office + bringing on Devendra Chaplot as advisor

PRODUCTS
10/ Voice Agents: build one in <60 min
11/ Work Agents: secure sandbox, Slack-native
12/ Content Studio: 50+ voices, voice cloning, motion-picture-grade dubbing
13/ Document Agents: extract, digitize & translate English and Indic
14/ Sarvam Code (beta): open-model coding harness

SOVEREIGN + DEVICES
15/ Chanakya: full on-prem stack built for defense & national security
16/ Kaze: AI-powered smart glasses
17/ Kivi: a voice dictation tool

[Shiva @LShivkumarreddy]

@SarvamAI glasses!😳

This is incredible! Congratulations 🥳

When will these hit the market?

[Nous Research @NousResearch]

Hermes Agent now runs Buzz.

The self-hostable workspace from @blocks puts humans and agents in the same messaging channels and codebase.

Three ways to use Buzz with Hermes (and vice versa):
- Buzz Desktop auto-discovers your Hermes install runs it locally
- A relay bridge gives it a hosted identity in your channels
- Connect via the Hermes Gateway to use Buzz as a full external platform with channels, DMs, threads, reactions, and cron delivery

https://t.co/srljIbERN7