🔄 In a Training Loop

Ashish Patnaik

PatnaikAshish

4 5 318

https://github.com/Ashish-Patnaik

AI & ML interests

Building small, efficient AI models for LLMs, Text-to-Speech, audio generation models, and little bit interest in automating stuff. Focused on low-resource training, model optimization, and practical open-source projects.

Recent Activity

liked a Space 4 days ago

prithivMLmods/FireRed-Image-Edit-1.0-Fast

liked a model 8 days ago

Egor3333/OpenVoice

upvoted an article 14 days ago

Building "Wall Street of AI Agents": How I Crammed a Multi-Agent Financial Simulation into a 1B Parameter Model

View all activity

Organizations

upvoted an article 14 days ago

Article

Building "Wall Street of AI Agents": How I Crammed a Multi-Agent Financial Simulation into a 1B Parameter Model

build-small-hackathon

•

17 days ago

• 1

upvoted a collection 17 days ago

PP-OCRv6

Collection

From 1.5M to 34.5M Parameters, Surpassing Billion-Scale VLMs on OCR Tasks • 19 items • Updated 14 days ago • 98

upvoted a collection 3 months ago

Raon

Collection

9 items • Updated May 21 • 46

upvoted a paper 3 months ago

Llama-Mimi: Speech Language Models with Interleaved Semantic and Acoustic Tokens

Paper • 2509.14882 • Published Sep 18, 2025 • 2

upvoted a collection 4 months ago

Qwen3 Voice Embedding

Collection

Standalone ECAPA-TDNN x-vector speaker encoders extracted from Qwen3-TTS. 1024-dim (0.6B) and 2048-dim (1.7B). • 4 items • Updated Feb 27 • 29

Ashish Patnaik

AI & ML interests

Recent Activity

Organizations

PatnaikAshish's activity

Building "Wall Street of AI Agents": How I Crammed a Multi-Agent Financial Simulation into a 1B Parameter Model