Wang Weiyi PRO
kaupane
AI & ML interests
None yet
Recent Activity
reacted to nwaughachukwuma's post with 👍 1 day ago
Can a text-only model + a vision toolkit (mm-ctx) match a native vision model?
We benchmarked 4 setups on 23 multimodal tasks (image, video, audio, PDF):
• glm-5.2 (text-only) + mm-ctx: 88.4
• gemini-3.5-flash (vision): 83
• deepseek-v4-pro (text-only) + mm-ctx: 79.4
• qwen3.6-35b-a3b (vision): 44.3
The best text-only setup closes 89% of the gap to the top vision model.
Full report: https://huggingface.co/blog/vlm-run/text-only-models-with-mm liked a model 7 days ago
moonshotai/Kimi-K3 liked a Space 9 days ago
OpenMOSS-Team/MOSS-transcribe-diarizeOrganizations
None yet
models 6
kaupane/ArtFlow
0.7B • Updated
kaupane/DiT-Wikiart-Large
Text-to-Image • 0.2B • Updated • 14 • 1
kaupane/DiT-Wikiart-Small
Text-to-Image • 22M • Updated • 6
kaupane/DiT-Wikiart-Base
Text-to-Image • 90.3M • Updated • 2
kaupane/ChessFormer-RL
Reinforcement Learning • 0.1B • Updated • 3 • 1
kaupane/ChessFormer-SL
Reinforcement Learning • 0.1B • Updated • 7 • 1
datasets 12
kaupane/AstroReason-Bench
Viewer • Updated • 56 • 118
kaupane/chess-positions
Viewer • Updated • 1.14B • 328
kaupane/nano-banana-pro-gen
Viewer • Updated • 1.25k • 98 • 10
kaupane/z-image-turbo-gen
Viewer • Updated • 50k • 198
kaupane/astro-reason
Updated • 28
kaupane/human-recaption
Viewer • Updated • 240k • 63 • 2
kaupane/vintage-photography-captions
Viewer • Updated • 445k • 56 • 2
kaupane/relaion-art-recap-zh
Viewer • Updated • 2.01M • 15 • 1
kaupane/artbench-captions
Viewer • Updated • 71.9k • 644 • 2
kaupane/magpie-qwen2.5-chinese-glm4.6-reasoning-coding
Viewer • Updated • 15.9k • 67 • 1