Marcello Seri

mseri

·

https://mseri.me

AI & ML interests

None yet

Recent Activity

upvoted a collection about 4 hours ago

liked a model 1 day ago

prism-ml/Ternary-Bonsai-27B-gguf

liked a model 1 day ago

thinkingmachines/Inkling

View all activity

Organizations

upvoted a collection about 4 hours ago

Luciole LLM

Open Source LLM in French, English, German, Spanish, Italian, Portuguese, Dutch and Arabic • 14 items • Updated 7 days ago • 14

upvoted a collection 7 days ago

DFlash

Block Diffusion for Flash Speculative Decoding • 23 items • Updated 18 days ago • 150

upvoted an article 15 days ago

Article

Why Specialization Is Inevitable

Dharma-AI

•

16 days ago

• 13

upvoted a collection 21 days ago

Ornith-1.0

Ornith-1.0 is a family of open-source LLMs specialized for agentic coding. • 8 items • Updated 20 days ago • 346

upvoted 5 collections about 1 month ago

PP-OCRv6

From 1.5M to 34.5M Parameters, Surpassing Billion-Scale VLMs on OCR Tasks • 19 items • Updated Jun 15 • 104

Mellum 2

Mellum2 model weights • 6 items • Updated Jun 1 • 126

Gemma 4 QAT

Gemma 4 QAT (Quantization-Aware Training) for 3x less memory use and near original accuracy. • 16 items • Updated 6 days ago • 105

Gemma 4 QAT Mobile

4 items • Updated Jun 5 • 46

Gemma 4 QAT Q4_0

19 items • Updated Jun 5 • 145

upvoted an article 3 months ago

Article

Granite 4.1 LLMs: How They’re Built

ibm-granite

•

Apr 29

• 83

upvoted 4 collections 3 months ago

DeepSeek-V4

6 items • Updated 20 days ago • 737

Ternary Bonsai

1.58-bit Bonsai models • 9 items • Updated Jun 4 • 100

Introspective Diffusion Language Models (I-DLM)

Model checkpoints for I-DLM. Paper: https://arxiv.org/abs/2604.11035 • 3 items • Updated Apr 14 • 11

Gemma 4

Gemma 4 is Google's new model family including including E2B, E4B, 26B-A4B, and 31B. • 36 items • Updated 6 days ago • 229

upvoted 2 collections 4 months ago

Gemma 4

15 items • Updated Jun 10 • 1.03k

Bonsai

1-bit Bonsai models • 7 items • Updated Jun 4 • 211

upvoted 4 collections 5 months ago

Unsloth Dynamic 2.0 Quants

New 2.0 version of our Dynamic GGUF + Quants. Dynamic 2.0 achieves superior accuracy & SOTA quantization performance. • 109 items • Updated about 15 hours ago • 761

Qwen3.5

Qwen3.5 is Qwen's new model family including Qwen3.5 Small: 0.8B, 2B, 4B, 9B and Qwen3.5 Medium: 35B-A3B, 27B, 122B-A10B and 397B-A17B. • 25 items • Updated 6 days ago • 163

Parakeet ASR

NeMo Parakeet ASR Models attain strong speech recognition accuracy while being efficient for inference. Available in CTC and RNN-Transducer variants. • 16 items • Updated about 3 hours ago • 79

Nemotron Speech

Open, state-of-the-art, production‑ready enterprise speech models from the NVIDIA Speech research team for ASR, TTS, Speaker Diarization and S2S • 13 items • Updated about 3 hours ago • 59