Gan Feng

vonjack

32 3 132

AI & ML interests

None yet

Recent Activity

updated a model about 5 hours ago

vonjack/hrm-text-agent-gguf

published a model about 6 hours ago

vonjack/hrm-text-agent-gguf

updated a model about 9 hours ago

vonjack/hrm-text-code-tools-sft

View all activity

Organizations

updated a model about 5 hours ago

vonjack/hrm-text-agent-gguf

1B • Updated about 5 hours ago

published a model about 6 hours ago

vonjack/hrm-text-agent-gguf

1B • Updated about 5 hours ago

updated a model about 9 hours ago

vonjack/hrm-text-code-tools-sft

Text Generation • 1B • Updated about 9 hours ago • 39

published a model about 9 hours ago

vonjack/hrm-text-code-tools-sft

Text Generation • 1B • Updated about 9 hours ago • 39

liked a model 10 days ago

0labs-in/Sky-3B-SORE

Text Generation • 3B • Updated May 22 • 101 • 2

liked a Space 11 days ago

DFM Mimir

🧠

Chat with the bilingual Danish-English Mimir model

New activity in jasoncarreira/hrm-text-code 11 days ago

Combining code & tool calling in a single model?

#1 opened 11 days ago by

vonjack

liked a model 11 days ago

jasoncarreira/hrm-text-agent

Text Generation • 1B • Updated Jun 5 • 36 • 5

liked 4 models 12 days ago

liked 2 models 2 months ago

mudler/Qwen3.6-35B-A3B-APEX-MTP-GGUF

36B • Updated 14 days ago • 28.3k • 71

baidu/Unlimited-OCR

Image-Text-to-Text • 3B • Updated Jul 29 • 3.11M • 4.16k

liked a model 4 months ago

z-lab/Qwen3.6-27B-DFlash

Text Generation • 2B • Updated Apr 27 • 157k • 397

liked a Space 5 months ago

Bonsai 1-bit WebGPU

🌳

217

Run 1-bit Bonsai LLMs locally in your browser on WebGPU

commented on The Optimal Architecture for Small Language Models 7 months ago

Tested on MPS with different block size & steps:

Benckmarking Dhara-70M | Device: mps:0 | Target: 50 tokens

Config | Time | Speed(TPS) | Accel | Output Preview

Original AR | 2.087s | 24.0 | 1.00x | The future of artificial intelligence is a big challenge. This world has the pot...
Diff (B1/S50) | 2.392s | 20.9 | 0.87x | The future of artificial intelligence is the most-first thing. This article was ...
Diff (B2/S25) | 1.162s | 43.0 | 1.80x | The future of artificial intelligence is the What. What1: The Future Future? ...
Diff (B5/S10) | 0.382s | 131.0 | 5.47x | The future of artificial intelligence is the This.,!!."."!!!!!!!!!!!!!!!!!!!!...
Diff (B10/S5) | 0.192s | 260.5 | 10.87x | The future of artificial intelligence is the !!!!!!!!!!!!!!!!!!!!!!!!!!!!!...

It seems that when we use AR or we use diffusion similar to AR (B1/S50), we have best quality. But it's slow.
When we make it faster, it will generate some bad results.