Nico Hezel
Neiko2002
AI & ML interests
None yet
Recent Activity
new activity about 16 hours ago
halt95/Qwen3.8-Flash-Next-W4A16-Merlin:TP2 and PP2 would be interesting liked a model 5 days ago
convaiinnovations/layaOrganizations
TP2 and PP2 would be interesting
10
#2 opened 16 days ago
by
Neiko2002
Support eagle3 spec decode with pipeline parallel- #50514
7
#1 opened 16 days ago
by
Neiko2002
4x 3090 support ?
5
#1 opened 18 days ago
by
Neiko2002
Issue: AttributeError when starting scripts/serve.sh with vision enabled.
2
#1 opened 16 days ago
by
Neiko2002
Amazing quant quality.
👍 2
3
#4 opened 24 days ago
by
Neiko2002
Restricted access to models
1
#1 opened about 1 month ago
by
Neiko2002
Amazing Quant
😎 1
6
#1 opened about 2 months ago
by
Neiko2002
Very fast and high quality quant
👍 5
2
#1 opened about 1 month ago
by
Neiko2002
Vision Tower problems
2
#1 opened about 1 month ago
by
Neiko2002
Benchmark results?
1
#1 opened about 2 months ago
by
Neiko2002
Thinking Loops, and Chat Template possible fix!
❤️ 4
20
#23 opened 2 months ago
by
danielcherubini
Much better with the original chat template
2
#3 opened 3 months ago
by
Neiko2002
Are there plans to release something around 9-12B in weight?
4
#3 opened 3 months ago
by
XuehangCang
Gets even better with the right template
🔥 1
#1 opened 3 months ago
by
Neiko2002
FP8 seems to be broken
➕ 1
3
#1 opened 3 months ago
by
Neiko2002
Original ninja.template provides better results
3
#1 opened 4 months ago
by
Neiko2002
comparison
17
#2 opened 5 months ago
by
kalle07
Broken config.json for vllm v0.21.0
#3 opened 4 months ago
by
Neiko2002
Improved quality by changing the chat_template.jinja
4
#1 opened 4 months ago
by
Neiko2002
tool calling?
1
#1 opened 5 months ago
by
Neiko2002