👋 Open to Work
Kaloyan Nikolov
sleepyeldrazi
·
AI & ML interests
ML engineer working on LLM inference and training. Finalizing my M.Sc. Computer Science @ RWTH Aachen.
Currently focusing on DGX Spark models/frameworks. Contact: konik98@gmail.com
Recent Activity
new activity 7 days ago
sleepyeldrazi/Ornith-1.5-REAP-20B-A3B-GGUF:List of pruned experts new activity 16 days ago
quimmedes/Qwen3.8-Flash-Next-MTP-GGUF:BF16 Missing currentlyOrganizations
None yet
List of pruned experts
5
#1 opened 12 days ago
by 
kdrapel
Early report running it on a DGX Spark, 4 concurrent @ 256k bf16 with SSD offload
👍 7
4
#31 opened 16 days ago
by sleepyeldrazi
BF16 Missing currently
1
#1 opened 16 days ago
by sleepyeldrazi
NVME vs Optane?
3
#29 opened 16 days ago
by 
AndyFaceHugger
Is it possible to offload n-gram to an NVMe SSD?
👀🚀 31
14
#11 opened 17 days ago
by 
lingyezhixing
Yet to see any gains from DFlash
👍 11
13
#3 opened 24 days ago
by 
Dnonmi
BenchMAXXXED
➕ 2
33
#28 opened about 2 months ago
by 
TawusGames
How can I run this well on an RTX 5070 ti?
8
#36 opened 29 days ago
by 
theonerm2
Text-only? Plans for multimodal release?
😔👀 4
2
#7 opened about 1 month ago
by sleepyeldrazi
Reproducible Laguna S 2.1 NVFP4 runaway reasoning: output budget scales 2K → 16K with 12/12 length exhaustion and no final answer
👀👍 8
7
#16 opened about 2 months ago
by 
darkmatter2222
Still looping after update - details inside
👀 2
22
#10 opened about 2 months ago
by 
cbert33
Model is decent, its faster than Qwen and pretty close on real world tool calls
❤️👀 3
7
#22 opened about 2 months ago
by 
Jakaryun
MOE or changing to A3B
3
#43 opened about 2 months ago
by 
Tom-Neverwinter
Awesome model
👍❤️ 16
11
#2 opened 3 months ago
by 
pengzhi27
Distilled Reasoning or Just XML Cosplay?
🔥 2
2
#2 opened 3 months ago
by 
SuperSonnix71
File path upper case issue
4
#8 opened 3 months ago
by sleepyeldrazi
File path case issue
#5 opened 3 months ago
by sleepyeldrazi
MTP Support
#7 opened 4 months ago
by sleepyeldrazi