https://
huggingface.co/tencent/Tencen
t-Hunyuan-Large
…
@reach_vb
-
Tencent Releases Hunyuan Large Language Model
By
–
-
Tencent Hunyuan Large 389B: New LLM Outperforms Llama DeepSeek
By
–
We're sooo back! – Tencent Hunyuan Large – 389B (Total) X 52B (Active) – beats Llama 3.1 405B, Mistral 8x22B, DeepSeek V2! Multilingual, 128K context, Utilizes GQA + CLA for KV Cache compression + Higher throughput Released Pre-train, Instruct & FP8 checkpoints on the Hugging
-
Invitation to discuss HF collaboration for model iterations
By
–
Thank you! I'd love to chat more and discuss how HF can help with futher iterations of the model, would you mind sending a DM!
-
350M Text-to-Speech Model Generates Impressive Coherent Audio
By
–
It's hilarious to see the model do go off the rails and just make random but coherent audio up – it's still quite impressive for a 350M Text to Speech model 👀 https://t.co/bqsULmSPlK pic.twitter.com/TMJWaHY2KS
— Vaibhav (VB) Srivastav (@reach_vb) 4 novembre 2024It's hilarious to see the model do go off the rails and just make random but coherent audio up – it's still quite impressive for a 350M Text to Speech model
-
LLaMa’s Transformative Impact: Open Source AI Leadership
By
–
LLaMa has had an unreal impact across all fields of AI!
— Vaibhav (VB) Srivastav (@reach_vb) 4 novembre 2024
Open Source is the way – This is how we win! https://t.co/bqsULmTnbiLLaMa has had an unreal impact across all fields of AI! Open Source is the way – This is how we win!
-
OuteTTS-0.1-350M: Compact Open-Source Voice Cloning Model
By
–
Smol TTS models are here! OuteTTS-0.1-350M – Zero shot voice cloning, built on LLaMa architecture, CC-BY license! 🔥
— Vaibhav (VB) Srivastav (@reach_vb) 4 novembre 2024
> Pure language modeling approach to TTS
> Zero-shot voice cloning
> LLaMa architecture w/ Audio tokens (WavTokenizer)
> BONUS: Works on-device w/ llama.cpp ⚡… pic.twitter.com/lXKAGwvvvHSmol TTS models are here! OuteTTS-0.1-350M – Zero shot voice cloning, built on LLaMa architecture, CC-BY license! > Pure language modeling approach to TTS
> Zero-shot voice cloning
> LLaMa architecture w/ Audio tokens (WavTokenizer)
> BONUS: Works on-device w/ llama.cpp -
AMD OLMo 1B: Efficient Open-Source Language Model Released
By
–
Smol models ftw! @AMD released AMD OLMo 1B – beats OpenELM, tiny llama on MT Bench, Alpaca Eval – Apache 2.0 licensed > Trained with 1.3 trillion (dolma 1.7) tokens on 16 nodes, each with 4 MI250 GPUs > Three checkpoints: – AMD OLMo 1B: Pre-trained model
– AMD OLMo 1B SFT: