Rishikesh (ऋषिकेश)
975 posts
@ai_rishikesh
LLM Researcher, Audio Generation, TTS & Image domain Open-Source enthusiast | Backend Dev | Photographer | Boston Celtics 🏀 | Man Utd FC ⚽️
🚀 Introducing the Qwen 3.5 Small Model Series Qwen3.5-0.8B · Qwen3.5-2B · Qwen3.5-4B · Qwen3.5-9B ✨ More intelligence, less compute. These small models are built on the same Qwen3.5 foundation — native multimodal, improved architecture, scaled RL: • 0.8B / 2B → tiny, fast, great for edge device • 4B → a surprisingly strong multimodal base for lightweight agents • 9B → compact, but already closing the gap with much larger models And yes — we’re also releasing the Base models as well. We hope this better supports research, experimentation, and real-world industrial innovation. Hugging Face: huggingface.co/collections/Qw… ModelScope: modelscope.cn/collections/Qw…
Introducing 𝑨𝒕𝒕𝒆𝒏𝒕𝒊𝒐𝒏 𝑹𝒆𝒔𝒊𝒅𝒖𝒂𝒍𝒔: Rethinking depth-wise aggregation. Residual connections have long relied on fixed, uniform accumulation. Inspired by the duality of time and depth, we introduce Attention Residuals, replacing standard depth-wise recurrence with learned, input-dependent attention over preceding layers. 🔹 Enables networks to selectively retrieve past representations, naturally mitigating dilution and hidden-state growth. 🔹 Introduces Block AttnRes, partitioning layers into compressed blocks to make cross-layer attention practical at scale. 🔹 Serves as an efficient drop-in replacement, demonstrating a 1.25x compute advantage with negligible (<2%) inference latency overhead. 🔹 Validated on the Kimi Linear architecture (48B total, 3B activated parameters), delivering consistent downstream performance gains. 🔗Full report: github.com/MoonshotAI/Att…
📢 Open-sourcing the Sarvam 30B and 105B models! Trained from scratch with all data, model research and inference optimisation done in-house, these models punch above their weight in most global benchmarks plus excel in Indian languages. Get the weights at Hugging Face and AIKosh. Thanks to the good folks at SGLang for day 0 support, vLLM support coming soon. Links, benchmark scores, examples, and more in our blog - sarvam.ai/blogs/sarvam-3…
What if your voice could speak every language? This pitcher just made it possible! 🌐🦈 #dubpro.ai Watch The New Episodes Of Shark Tank India Season 5, Streaming Now Mon-Fri, 8 PM on Sony LIV. Watch Free On Mobile Only. @AnupamMittal @amangupta0303 @namitathapar @vineetasng
Qwen3-TTS is officially live. We’ve open-sourced the full family—VoiceDesign, CustomVoice, and Base—bringing high quality to the open community. - 5 models (0.6B & 1.8B) - Free-form voice design & cloning - Support for 10 languages - SOTA 12Hz tokenizer for high compression - Full fine-tuning support - SOTA performance We believe this is arguably the most disruptive release in open-source TTS yet. Go ahead, break it and build something cool. 🚀 Everything is out now—weights, code, and paper. Enjoy. 🧵 Github: github.com/QwenLM/Qwen3-T… Hugging Face: huggingface.co/collections/Qw… ModelScope: modelscope.cn/collections/Qw… Blog: qwen.ai/blog?id=qwen3t… Paper: github.com/QwenLM/Qwen3-T… Hugging Face Demo: huggingface.co/spaces/Qwen/Qw… ModelScope Demo: modelscope.cn/studios/Qwen/Q… API: alibabacloud.com/help/en/model-…
🚨 Chatterbox Turbo is now live on fal! 🗣️ Ultra-fast, open-source text-to-speech built for real-time voice AI ⚡ Up to 6× faster-than-real-time 🎭 Paralinguistic tags for non-verbal reactions: [sigh], [chuckle], [laugh], [gasp] + more 🎙️ Instant voice cloning from ~5 seconds of audio. Reactions stay in the same voice
VoxCPM Technical Report is here! ⚡️ We’re taking realistic speech generation to the next level of Efficiency. 📉 ✨ Highlights: 🚫 Tokenizer-Free: Pure end-to-end continuous modeling for high fidelity. 🧠 Hierarchical Design: TSLM + RALM ensures both stability & expressivity. ⚡ Blazing Fast: Achieves an RTF as low as 0.17 on consumer GPUs. Dive into the tech that makes it possible: 🔗 Technical Report: arxiv.org/abs/2509.24650 🤗 Model: huggingface.co/openbmb/VoxCPM… 🎮 Demo: huggingface.co/spaces/openbmb… #AI #TTS #OpenBMB #VoxCPM #OpenSource
Your clients and prospects are already talking to AI. Why aren’t they talking to yours? 🤔 That changes today with @myclone_is - your knowledge, your clients, your AI. Let me tell the quiet part out loud: for consultants, advisors, and coaches, you are the product. We spent 4 months with 450+ consultants & coaches and captured it in a 2-min film about Jo. Who is ready to use AI to scale themselves? #MyClone
