Nanbeige LLM Lab

Nanbeige's activity

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

leran1995 

in Nanbeige/Nanbeige4.2-3B about 6 hours ago

Awesome!

👍 1

1

#24 opened about 9 hours ago by

AsThirtyThree

leran1995 

updated a model 3 days ago

Nanbeige/Nanbeige4.2-3B

Text Generation • 4B • Updated 3 days ago • 26.9k • 594

leran1995 

in Nanbeige/Nanbeige4.2-3B 3 days ago

Amazing for its size, but spirals into questionable solutions once it fails something

1

#21 opened 3 days ago by

Mk2Oracle

flust 

updated a model 5 days ago

Nanbeige/Nanbeige4.2-3B

Text Generation • 4B • Updated 3 days ago • 26.9k • 594

leran1995 

in Nanbeige/Nanbeige4.2-3B 5 days ago

Massive activation at layer 21, and three follow-ups on loop depth + LoopSplit

1

#20 opened 5 days ago by

an0nya

Suitability for 8 GB graphics cards

👀 4

3

#18 opened 5 days ago by

thermi6

Undocumented config params + two questions on loop depth and length-control RL

1

#16 opened 6 days ago by

an0nya

leran1995 

updated a model 5 days ago

Nanbeige/Nanbeige4.2-3B-Base

Text Generation • 4B • Updated 5 days ago • 2.4k • 50

leran1995 

in Nanbeige/Nanbeige4.2-3B 6 days ago

Some questions about training

1

#14 opened 7 days ago by

mobinx

leran1995 

in Nanbeige/Nanbeige4.2-3B-Base 7 days ago

Request for Access to the Pre-training Dataset for Nanbeige/Nanbeige4.2-3B-Base (Open-Source Research)

👍 1

2

#1 opened 7 days ago by

arpitsh018

leran1995 

in Nanbeige/Nanbeige4.2-3B 7 days ago

the modal is tiny but kv cache exploding!

👍 6

2

#10 opened 9 days ago by

rosspanda0

LiZongqiang 

updated a model 9 days ago

Nanbeige/Nanbeige4.2-3B

Text Generation • 4B • Updated 3 days ago • 26.9k • 594

leran1995 

in Nanbeige/Nanbeige4.2-3B 9 days ago

preserve_thinking

5

#9 opened 9 days ago by

owao

Please specify max context size in model's card.

1

#8 opened 9 days ago by

Reverger

Any plans for FP8?

2

#5 opened 9 days ago by

spanspek

Add community evaluation results for CLAW-EVAL, GPQA, HLE, HMMT_FEB_2026, SWE-BENCH_PRO, SWE-BENCH_VERIFIED, TERMINAL-BENCH-2.0

#7 opened 9 days ago by

nielsr

"Weimplify is asked:"

4

#4 opened 10 days ago by

owao

leran1995 

in Nanbeige/Nanbeige4.2-3B 10 days ago

Q on the model

4

#3 opened 10 days ago by

TomLucidor

gguf

➕ 5

4

#1 opened 10 days ago by

Tapka

damn i was in the process of doing something very similar

1

#2 opened 10 days ago by

nraxl1