specimba

specimba's activity

• All
• Models
• Datasets
• Spaces
• Buckets
• Papers
• Collections
• Community
• Posts
• Upvotes
• Likes
• Articles

New activity in build-small-hackathon/NEXUS-Visual-Weaver 1 day ago

Upgrade to FLUX.2-klein-4B (Apache, 4B, sub-second) via Flux2KleinPipeline

#2 opened 1 day ago by

specimba

Fix ZeroGPU crash: import spaces first, real @spaces.GPU, FLUX.1-schnell

#1 opened 1 day ago by

specimba

New activity in Trendyol/Trendyol-TTS 1 day ago

HF Space Demo

❤️ 3

4

#1 opened 1 day ago by

atasoglu

New activity in specimba/nexus-os-v2-dualmode 2 days ago

Subject: Collaboration / Joining Xerv: Building Qubik (5B Edge Search Brain)

👀 1

3

#1 opened 2 days ago by

Phase-Technologies

commented a paper 3 days ago

Reinforcement Learning via Self-Distillation

Paper • 2601.20802 • Published Jan 28 • 49 •

6

New activity in build-small-hackathon/Darwin-35B-A3B-Opus 4 days ago

Fix Darwin 35B load: CPU offload flag + spaces pkg + persistent cache

#6 opened 4 days ago by

specimba

Fix: add spaces dependency + force rebuild

#5 opened 4 days ago by

specimba

Fix: Move model load inside @spaces.GPU to prevent CPU OOM crash

#4 opened 5 days ago by

specimba

New activity in K0D3IN/gemma-3-1B-it-heretic 6 days ago

keep going on

1

#1 opened 6 days ago by

specimba

New activity in prism-ml/bonsai-image-binary-4B-mlx-1bit 6 days ago

Hmm, pretty fast on RTX 3090 24 GB VRAM, though why censored?

2

#1 opened 6 days ago by

JLouisBiz

New activity in cb1c7/minicpm5-twitch-chat-style-space 6 days ago

need intelligence

1

#1 opened 6 days ago by

specimba

New activity in prism-ml/Bonsai-Image-Demo 11 days ago

Impressive, but...

1

#1 opened 11 days ago by

Gershwin69

New activity in huihui-ai/Huihui-gemma-4-E4B-it-abliterated 13 days ago

Model Repack

3

#1 opened 13 days ago by

Matthew3179

New activity in build-small-hackathon/Darwin-35B-A3B-Opus 15 days ago

PR #2 requirements: Remove llama.cpp, add transformers and full ML stack

1

#3 opened 16 days ago by

specimba

PR #2: Replace llama.cpp with transformers + spaces.GPU + temp grid

1

#2 opened 16 days ago by

specimba

New activity in build-small-hackathon/Darwin-35B-A3B-Opus 16 days ago

Fix libcudart.so.12: force CPU-only llama-cpp build

1

#1 opened 16 days ago by

specimba

commented a paper 16 days ago

Darwin Family: MRI-Trust-Weighted Evolutionary Merging for Training-Free Scaling of Language-Model Reasoning

Paper • 2605.14386 • Published 24 days ago • 60 •

5

New activity in specimba/nexus-os-lab 17 days ago

Apply for a GPU community grant: Academic project

🚀 1

#1 opened 17 days ago by

specimba

commented a paper 27 days ago

DFlash: Block Diffusion for Flash Speculative Decoding

Paper • 2602.06036 • Published Feb 5 • 83 •

7

New activity in huggingface/InferenceSupport 27 days ago

RedHatAI/Qwen3-8B-speculator.dflash

🚀 1

#9802 opened 27 days ago by

specimba