Wladastic
Wladastic
AI & ML interests
LLM Agents
Recent Activity
liked a model 5 days ago
unsloth/GLM-5.3-Flash-GGUF liked a model 5 days ago
AtomicChat/Qwen3.8-Flash-Next-GGUF liked a model 5 days ago
Qwen/Qwen3.8-Flash-NextOrganizations
None yet
VRAM on v3
10
#1 opened 21 days ago
by
Oxidez
Try Qwen3.8-27B in Unsloth Desktop with Thinking toggles!
🔥 11
14
#23 opened 18 days ago
by
danielhanchen
DeepSeek-V4-Flash-0731-Ombro-v1 performance.
6
#1 opened 26 days ago
by
Oxidez
text encoder int4 model has a few extra bytes
👍 1
1
#5 opened 28 days ago
by
Wladastic
How did you quantize without imatrix?
#27 opened 29 days ago
by
Wladastic
I think I've pinned down good sampler settings! Someone please sanity check
13
#16 opened about 1 month ago
by
pegasus912
Upstream the `laguna` branch?
➕ 1
2
#19 opened about 1 month ago
by
coder543
MXFP4-Moe Quant please
#4 opened about 1 month ago
by
Wladastic
Thinking Loops, and Chat Template possible fix!
❤️ 4
20
#23 opened about 1 month ago
by
danielcherubini
MXFP4-Moe?
#13 opened about 1 month ago
by
Wladastic
Great on HERMES! Noob Question
➕ 2
4
#8 opened 2 months ago
by
GravyLuver
Misunderstood what it is for, but it works super good with Openclaw, wtf?!
👍 8
1
#9 opened 2 months ago
by
Wladastic
Same with QAT?
🚀 3
3
#9 opened 2 months ago
by
Wladastic
model.safetensors file size :939 MB
4
#1 opened 3 months ago
by
devops724
Seperate mmproj file?
➕ 4
1
#14 opened 3 months ago
by
Wladastic
IQ4_XS cache quantization benchmarks
2
#27 opened 4 months ago
by
MortiDahlaine
Managed to squeeze more efficiency out of this
🔥 2
#20 opened 4 months ago
by
Wladastic
🎉 ComfyUI ready
👍 8
5
#3 opened 5 months ago
by
drbaph
Best quality! Using this onnx in my webui api
#2 opened 4 months ago
by
Wladastic