Ilya Pereverzin
NodeLinker
AI & ML interests
Isn't it amazing that we let a computer think like a human?
Recent Activity
liked
a model 4 days ago
Qwen/Qwen3.5-35B-A3B-Base liked
a model 4 days ago
Qwen/Qwen3.5-122B-A10B liked
a model 4 days ago
Qwen/Qwen3.5-35B-A3B Organizations
Dark theme readability of benchmarks is pooor
β 9
1
#4 opened 12 days ago
by
RASMUS
Fix README accessibility: enable adaptive light/dark theme for tables
#8 opened 12 days ago
by
NodeLinker
Yo, the model is just AMAZING!
#1 opened 19 days ago
by
NodeLinker
Question about Step 3.5 Flash Base model weights release
π 1
3
#21 opened 22 days ago
by
NodeLinker
[Urgent Suggestion] Complete Deployment Guide: SDPA Patch (2x speed), 4-bit Fix for 8GB GPUs & Visual Examples
1
#6 opened 23 days ago
by
NodeLinker
Fixing line break & Critical feedback on missing features (OD/Segmentation)
11
#5 opened 24 days ago
by
NodeLinker
Fixing a line break error in Python.
#4 opened 24 days ago
by
NodeLinker
Questions regarding LoRA compatibility (Base vs Turbo) and Ecosystem support
#1 opened 25 days ago
by
NodeLinker
Congratulations on this release and on one important realization!
π₯ π 19
5
#5 opened about 1 month ago
by
Dampfinchen
Is a dedicated Tech Report planned for GLM-4.7-Flash?
1
#8 opened about 1 month ago
by
NodeLinker
Wrong precision at text encoder qwen_3_4b_fp4_flux2.safetensors
#1 opened about 1 month ago
by
NodeLinker
Status of 32B and 235B models mentioned in the report?
π€ 3
#1 opened 2 months ago
by
NodeLinker
Lvl 6-8 Real solution
#1 opened 2 months ago
by
NodeLinker
inclusionAI/LLaDA2.0-mini
π π₯ 1
#6938 opened 2 months ago
by
NodeLinker
inclusionAI/LLaDA2.0-mini
π π€ 3
1
#6372 opened 3 months ago
by
llmq1
[DELETE ME]
#5 opened 3 months ago
by
NodeLinker
[DELETE ME]
#45 opened 3 months ago
by
NodeLinker
Width and Height are swapped
π₯ 1
2
#5 opened 3 months ago
by
furaidosu
Eval request
π π₯ 2
3
#416 opened 4 months ago
by
Pentium95
Π·Π΄ΡΠ°ΡΡΠ²ΡΠΉΡΠ΅
7
#6 opened 6 months ago
by
Tim324325