Tech Entrepreneur
Madurai
Joined September 2021
- Tweets392
- Following2.5K
- Followers535
- Likes15K
Pinned Tweet
With @bot ’s help, I’ve launched a Peer to Peer ZDR Compute network with models at half their price compared to OpenRouter. The reason we are able to offer this is we don’t incur DataCenter costs the cosmic salaries of Valley Network engineers. Our providers are SLA tested gaming rigs and unsold DC inventory. We carry @deepseek_ai V4.1 Flash 1017 Uncensored, @Zai_org GLM 5.3 Flash Uncensored and @Alibaba_Qwen 3.8 27B Uncensored. Immediate access to our uncensored models and you can pay via Fiat, Crypto or UPI
I’m a 1 person startup so please help me out by Liking & Reposting this and collect $10 or Rs.1000 worth coupons in your DMs, Cheers!
Today is such a good day, an indic open weights and in this case a fully opensource AI model was launched @jiteshkprasad congratulations to you and your team, leading by example
Vemana is out. It's a 278M Telugu model we trained from scratch, live, on one GPU, for about $69.
And I think the result surprised even me: on new Telugu Wikipedia pages it beats every open base model under 1B that we tested, including nanochat-telugu-560M, which is twice its size.
The weights, every checkpoint and the eval code are public. And you can run it right in your browser, on your own device, nothing gets sent anywhere.
This is a very expressive Indic language TTS that can run in your own premises. Thank you for making this open weights @Dheemanthredy , we know the substantial amount of work you and your team would’ve poured into it. Good day for Indian AI Scene. Also kudos on your HF page to honestly say which base model your post trained this on. This transparency is very much needed from other players too
Maya-2 voice model weights are officially live! 🇮🇳🇮🇳
The frontier voice model for India is now completely open-weight.
• Outperforms ElevenLabs v3, Sarvam Bulbul v3, and Cartesia Sonic 3.5
• 1.5B+ people covered across all 28 states and 8 union territories
• SOTA voice synthesis across 11 official languages
• 20 native speakers pre-loaded, ready out of the box
• Best-in-class zero-shot voice cloning
Built by a team of five from the land of languages.
Go build and have fun!
Next air crash story loading with the scientist all together heading for a conference/prize ceremony
1 person, 1 week, 1 @ASUS ProArt P16 laptop. Built entirely using @OpenAI Astra and Sol 6.1
in @UnrealEngine using free @fab assets.
Game ETA - 3 days, will be opensourced #BuildWithAI
Sorry, forgot to add that I’ve never worked with Blender or Unreal or ever made any game so far. The only mod I tweaked was a bat mobile for @NeedforSpeed 3 Hot Pursuit
Here’s Abliterated @Alibaba_Qwen 3.8 27B BF16 and various quants:
huggingface.co/junafinity/Qw…
Our group of researchers along with an autonomous AI Agent Research Team have found links and are using #Keezhadi excavation findings to help translate Indus Valley writings, some of which have been deciphered as proto-dravidian Tamil: huggingface.co/blog/junafini…
Reflexive refusals down 88%
@googlegemma 4 12B is now ready to be your red team - #abliterated
Thanks to @osmAPI_off @TervPro @TervProTech
huggingface.co/blog/junafini…
If you’ve got 128GB unified RAM on any Mac, here’s the 106GB Step 3.7 Flash for you:
huggingface.co/osmapi/Step-3…
MXFP4 MLX version is uploading
We liberated @claudeai opus distilled @Alibaba_Qwen 3.6 27B by @KyleHessling1 & Jackrong with Heretic abliteration tool kit. Quantized models with vision preserved:
huggingface.co/osmapi/osmQwo…
huggingface.co/osmapi/osmQwo…
huggingface.co/osmapi/osmQwo…
Thanks to @osmAPI_off @TervPro 's student research team
Arjun Reddy retweeted
A must attend webinar on what to build with our SoTA speech recognition model and the recent upgrades on diarisation and accuracy.
We're hosting a webinar on Saaras V3, our speech-to-text model for teams building Voice AI for India.
Date: Thursday, May 21
Time: 5:00 PM IST
In this session, we'll break down what it takes to ship Voice AI that works reliably across noisy environments, regional accents, mixed-language speech, and multiple speakers.
Register here: links.sarvam.io/speech-to-te…
