Cosmo

cosmojg

·

https://cosmo.red

AI & ML interests

Machine learning and computational neuroscience

Recent Activity

liked a model 2 days ago

moondream/moondream3.1-9B-A2B

liked a model 2 days ago

thinkingmachines/Inkling

liked a dataset 24 days ago

futo-org/swipe.futo.org

View all activity

Organizations

None yet

upvoted 2 collections 3 months ago

talkie-13b

talkie-1930-13b is a vintage language model trained on pre-1931 English-language text. See https://github.com/talkie-lm/talkie to run talkie. • 3 items • Updated Apr 21 • 57

DFlash

Block Diffusion for Flash Speculative Decoding • 23 items • Updated 20 days ago • 150

upvoted a collection 4 months ago

Gemma 4

Gemma 4 is Google's new model family including including E2B, E4B, 26B-A4B, and 31B. • 43 items • Updated about 12 hours ago • 239

upvoted an article 4 months ago

Article

KV Caching Explained: Optimizing Transformer Inference Efficiency

not-lain

•

Jan 30, 2025

• 372

upvoted a collection 4 months ago

Gemma 4

15 items • Updated Jun 10 • 1.04k

upvoted a collection 5 months ago

Qwen3.5

21 items • Updated Mar 9 • 1.72k

upvoted a paper 5 months ago

Block Diffusion: Interpolating Between Autoregressive and Diffusion Language Models

Paper • 2503.09573 • Published Mar 12, 2025 • 77

upvoted a collection 5 months ago

BD3-LMs

https://m-arriola.com/bd3lms/ • 4 items • Updated Apr 11 • 31

upvoted 2 collections 6 months ago

Trinity-Large

8 items • Updated Mar 30 • 42

Qwen3-TTS

7 items • Updated Jan 22 • 372

upvoted a paper 7 months ago

Diffusion Knows Transparency: Repurposing Video Diffusion for Transparent Object Depth and Normal Estimation

Paper • 2512.23705 • Published Dec 29, 2025 • 45

upvoted 6 collections 7 months ago

Olmo 3.1

The latest members of the Olmo 3 family: another 3 weeks of RL for 32B Think, the 32B Instruct model, large post-training research datasets... • 9 items • Updated Dec 23, 2025 • 54

Molmo2

Artifacts for the Molmo2 release • 5 items • Updated Mar 2 • 37

Bolmo

Artifacts for the Bolmo release: https://allenai.org/papers/bolmo. • 4 items • Updated Dec 23, 2025 • 12

Jan-v2-VL

Jan-v2-VL: a family of VLM focused on reliable, many-step task execution. • 9 items • Updated Mar 13 • 40

Mistral Large 3

A state-of-the-art, open-weight, general-purpose multimodal model with a granular Mixture-of-Experts architecture. • 4 items • Updated Dec 2, 2025 • 101

Ministral 3

A collection of edge models, with Base, Instruct and Reasoning variants, in 3 different sizes: 3B, 8B and 14B. All with vision capabilities. • 9 items • Updated Dec 2, 2025 • 171

upvoted a collection 8 months ago

FLUX.2

Our second generation of FLUX • 21 items • Updated Apr 6 • 258

upvoted 2 collections 10 months ago

Granite 4.0 Language Models

Efficient language models for multilingual generation, coding, RAG, and AI assistant workflows. • 11 items • Updated Apr 29 • 220

Granite Quantized Models

Quantized versions of IBM Granite models. • 50 items • Updated 18 days ago • 36