Josh Warner
JDWarner
ยท
AI & ML interests
None yet
Recent Activity
new activity 41 minutes ago
nvidia/Nemotron-Cascade-2-30B-A3B:pruned version liked a model 4 days ago
scottgl/Qwen3.5-122B-A10B-MTP-NVFP4 liked a model 4 days ago
Multilingual-Multimodal-NLP/IndustrialCoderOrganizations
None yet
pruned version
๐ฅ๐ 1
1
#16 opened about 13 hours ago
by
pirola
There's got to be a better way.
23
#6 opened 5 days ago
by
phil111
Recall from embed documents not as good as the original
5
#4 opened 7 days ago
by
o0Linny0o
A wild idea / suggestion...
๐ฅ 3
2
#4 opened 13 days ago
by
MrDevolver
Consider releasing full BF16 weights
2
#1 opened 13 days ago
by
JDWarner
good model
5
#1 opened 14 days ago
by
Roman1111111
Work great on 3090 except for weird (...) generation
โค๏ธ 1
5
#1 opened 14 days ago
by
ortegaalfredo
Qwopus with visual capabilities?
2
#19 opened 15 days ago
by
AQLabs
Does it improve or not?
๐ฅ 2
14
#7 opened 19 days ago
by
ThePabli
Security/Compliance Audit: EU AI Act & NIST Exposure
๐ฅ 1
3
#8 opened 18 days ago
by
tradeapollo
FP8 models
3
#1 opened 19 days ago
by
ecopoiesis
IQ5_K 136.891 GiB
๐ฅ 2
30
#9 opened about 2 months ago
by
Hunterx
Request: GGUF / quantized weights for Intern-S1-Pro
1
#7 opened about 2 months ago
by
gileneo
INT8 quantization for KVCache on DGX Spark/GB10
3
#6 opened about 2 months ago
by
JDWarner
This just trades general performance for domain specific gains.
๐ฅ๐ 16
11
#3 opened 7 months ago
by
phil111
Disable thinking mode in Jan-v1-4B model
2
#9 opened 7 months ago
by
vuhaix95