Solopreneur | indie hacker
Kochi, Kerala
Joined April 2010
- Tweets6K
- Following1.4K
- Followers307
- Likes5.1K
Krishna Sangeeth KS retweeted
An early attempt at getting @googlegemma 's Embedding Gemma 2 to paint using its shared understanding of text and image.
30 ms(around 30 images per second) per image on the Apple Neural Engine on a base Mac M1 pro. generates an image at the end of every word. Trained for about 12 hours on an A100. It is still undertrained. But, with enough data, images can get a lot better.
Krishna Sangeeth KS retweeted
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!
Krishna Sangeeth KS retweeted
Unlimited OCR 2 is here, and it's open source. 📖 ✨
Longer documents can mean a lot more attention computation. Our new SR-SWA mechanism keeps that in check by condensing text-visual attention interactions into a fixed-size set of working references.
Here's what it delivers:
- Document parsing at virtually any length, within a finite context window
- Nearly constant per-step decoding latency and memory footprint as documents grow longer
- State-of-the-art performance on MPDocBench and MarathonDocBench, covering documents from dozens to hundreds of pages
Check out the model here ↓
Krishna Sangeeth KS retweeted
I regret to inform you that nobody cares about your potential. Potential is nothing without execution. When you're young, everyone cares about your potential. You get used to focusing on it. But as you get older, no one cares about your potential. Just your delivery against it.
Krishna Sangeeth KS retweeted
This is what Crome saying “this is your last opportunity to fix the mess you’ve created, else you’re out.”
Pretty clear they will be out.
We’re talking about an eng team unable or unwilling to run a RCA and publish it, when everyone assumed they could
bugzilla.mozilla.org/show_bu…
Krishna Sangeeth KS retweeted
Can you use EmbeddingGemma 2 for Speech-to-Text?
This experiment uses EmbeddingGemma 2’s unified audio-text vector space to match speech against text via vector similarity.
The result? Transcribing an hour of audio in about 13 seconds, locally!
Krishna Sangeeth KS retweeted
Effective Immediately, all organizations using PHP for their systems are advised to be prepared for an update addressing two critical vulnerabilities. PHP is widely used for web applications. These vulnerabilities often provide a direct path to the server's file system or database. If an immediate update is not feasible, it is recommend that organizations take the system offline until it can be fully mitigated. NSA will follow up shortly with more specific guidance.
Krishna Sangeeth KS retweeted
🚀 Introducing DeepInfer, our in-house inference framework, now fully open source!
Powering DeepSeek-V4.1-Flash at scale with high throughput and low latency. We’re sharing our production stack to help the community lower compute costs, reduce energy use, and make AI more affordable.
🔗 github.com/deepseek-ai/DeepI…
1/4
Krishna Sangeeth KS retweeted
Most of these tricks are not specific to speech and should carry over to other transformers. The full recipe, with ablations, is in the blog post: kyutai.org/blog/2026-10-20-p…
Krishna Sangeeth KS retweeted
Pocket TTS has been getting updates those last few days. While it was ~100M parameters, we release smaller versions, down to 1.9M: an 18.2M model with the quality of the original, a 6.9M model, and a 1.9M model that runs 103x faster than real time on one CPU core. All with voice cloning.
Krishna Sangeeth KS retweeted
Over the past month, we completely switched to frontier LLMs to control a robot in a customer deployment.
By doing this, we have a classic System 1/System 2 architecture, but on steroids.
For the high-level reasoning layer, we use an agent based on Opus and Astra, which are many generations more capable than any VLA. For the fast execution layer, we use computer vision code written by the model, which is fast, editable, and reliable.
As a bonus, we got in-context learning, which is just a basic feature of LLMs, and continual learning, since whenever a new situation arises, the large model steps in and adapts the pipeline.
And it just feels better to be on the right side of the Bitter Lesson, benefiting from the billions being invested in frontier models around the world.
Krishna Sangeeth KS retweeted
This week, we launched Indra, a 1M parameter OCR model that's 15x faster than Tesseract on CPU and much more accurate.
I still can't believe how good this model is.
Krishna Sangeeth KS retweeted
FLUX 3 Image Klein is here.
Open Weights with the quality of a large frontier model. Customizable and fast.
5.6B parameters and runs on a single consumer GPU.
Native 4K, bounding boxes, and pixel-perfect editing.
Krishna Sangeeth KS retweeted
Meet Mellum2.1 – JetBrains' open-source 12B MoE model built for coding agents and local execution.
🚀 140+ T/S in local testing scenarios
🇪🇺 Ethically trained European open model
💻 Efficient worker model inside IDE agent workflows
Try it locally: huggingface.co/collections/J…
Krishna Sangeeth KS retweeted
We’re releasing the world’s best Indic voice AI model, fully open-weight. 🇮🇳🇮🇳
With over 1.5 billion people and 22 official languages, the future of AI for India is voice-first.
Maya-2 officially outperforms ElevenLabs v3, Cartesia Sonic 3.5, and Bulbul v3 across Indian languages.
It will be completely free for every builder very soon.
Krishna Sangeeth KS retweeted
How it works: the model stays untouched. A small reader on top takes each slice of audio, finds the nearest of ~229,000 words in the model's own text space, and strings them together. It trained in under two hours on 8×H100, on openly licensed speech.
On the same base M3 Air, it runs about twice as fast as IBM's Granite Speech TurboCTC. The table has more Macs and models.
Its word error rate is about 5% right now, against 2–3% for leading STT models. The next iteration is to train a better word reader to improve that.
Here is a free Mac app you can download to test it.
haku-models.tslfiles.org/eg2…
Here's google's post
xitter.cf/googlegemma/status/210…
Introducing EmbeddingGemma 2! 🚀
Our lightweight, multimodal embedding model maps text, code, images, video, and audio into a single, unified embedding space. Optimized for on-device use cases, it features:
- 740M parameter form factor with modular encoders
- Flexible dimension sizes (768dim-128dim) via Matryoshka Representation Learning (MRL)
- 8K context window (4x larger than text-only EmbeddingGemma)
- A commercially permissive Apache 2.0 license
Krishna Sangeeth KS retweeted
For many of us journalists, John Brittas has become over the years one of the first people we call when we face harassment from the state. He has stood up for journalists and others facing oppression, and has been a strong voice of the Opposition.
Today, he was dragged into a police bus while protesting, assaulted, and dumped on the roadside.
Solidarity with you @JohnBrittas. If this is how the police treat an elected MP, imagine what an ordinary citizen faces.
