John Leimgruber III PRO

ubergarm's activity

• All
• Models
• Datasets
• Spaces
• Buckets
• Papers
• Collections
• Community
• Posts
• Upvotes
• Likes
• Articles

New activity in RDson/Qwen3.6-27B-MTP-Q4_K_M-GGUF about 8 hours ago

提速效果不理想

7

#1 opened 6 days ago by

androidli

New activity in Lorbus/Qwen3.6-27B-int4-AutoRound about 8 hours ago

How does this fair against other quants without MTP like unsloth?

👍 1

1

#4 opened 11 days ago by

Crigges

New activity in ubergarm/Qwen3.6-27B-GGUF about 11 hours ago

How to use MTP in GGUF?

16

#2 opened 6 days ago by

Friedland

New activity in ubergarm/Qwen3.6-27B-GGUF 5 days ago

Q6_0 use over Q6_K?

2

#3 opened 5 days ago by

resynth

New activity in ubergarm/Qwen3.6-27B-GGUF 7 days ago

Great model for single GPU use cases.

🔥 4

16

#1 opened 11 days ago by

phakio

New activity in ubergarm/Qwen3.5-122B-A10B-GGUF 9 days ago

How to split this model between 2 (3) GPUs and CPU/RAM ?

30

#12 opened about 2 months ago by

mancub

New activity in ubergarm/Qwen3.5-35B-A3B-GGUF 11 days ago

Any plans to release Qwen3.6 ?

1

#3 opened 12 days ago by

ittokasso

New activity in ubergarm/Qwen3.5-27B-GGUF 12 days ago

smol-IQ4_NL with turboquant is outstanding for 24GB VRAM

🔥 1

3

#7 opened 12 days ago by

IHaveNoClueAndIMustPost

New activity in ubergarm/Kimi-K2.6-GGUF 12 days ago

Much faster IQ2_KS quantization PPL results?

3

#10 opened 13 days ago by

gghfez

really awesome speeds! running at 256k context.

🔥 1

5

#11 opened 13 days ago by

mtcl

Is IQ3_K mainline compatible?

1

#12 opened 12 days ago by

TimothyRoo

Testing IQ3_K

🔥 1

8

#6 opened 14 days ago by

shewin

How to disable thinking?

4

#9 opened 13 days ago by

Hansi2024

New activity in ubergarm/Kimi-K2.6-GGUF 13 days ago

How to use mmproj

👍 1

1

#8 opened 13 days ago by

curiouspp8

Testing smol-IQ3_KS

2

#7 opened 14 days ago by

shewin

No think tags.

10

#4 opened 14 days ago by

DrRos

New activity in ubergarm/Kimi-K2.6-GGUF 14 days ago

smol-IQ2_KL bench, GPUs

🔥 1

5

#5 opened 14 days ago by

curiouspp8

New activity in unsloth/Kimi-K2.6-GGUF 14 days ago

Q4_0 vs native INT4 QAT fidelity

👍 1

6

#4 opened 15 days ago by

SpacetimeAI

What kind of Q4_0 are you using for ffn_(gate|up|down)_exps?

🔥👍 2

2

#2 opened 15 days ago by

ubergarm

New activity in ubergarm/GLM-5.1-GGUF 14 days ago

GLM 5.1 vs GLM 5 - burns A LOT output tokens on thinking

👍 1

35

#6 opened 25 days ago by

curiouspp8