Models that run well on a *standalone* RTX a6000's 48gb of VRAM.
Ben Kelly PRO
YellowjacketGames
AI & ML interests
None yet
Recent Activity
upvoted
a
changelog
about 6 hours ago
Sort Models by Parameter Size
published
an
article
1 day ago
Flux 2 Dev on a single RTX A6000 48gb GPU #
updated
a collection
1 day ago
RTX a6000 48gb
Organizations
None yet
-
unsloth/Nemotron-3-Nano-30B-A3B-GGUF
Text Generation • 32B • Updated • 102k • 229 -
unsloth/GLM-4.7-Flash-GGUF
Text Generation • 30B • Updated • 112k • 235 -
black-forest-labs/FLUX.2-dev
Image-to-Image • Updated • 106k • • 1.27k -
unsloth/Llama-4-Scout-17B-16E-Instruct-GGUF
Image-to-Text • 108B • Updated • 21k • 129
TPS can be as low as 1.0, seriously. its SLOW.
-
unsloth/GLM-4.7-GGUF
Text Generation • 358B • Updated • 130k • 179 -
unsloth/DeepSeek-R1-0528-GGUF
Text Generation • 671B • Updated • 4.44k • 193 -
unsloth/Llama-4-Maverick-17B-128E-Instruct-GGUF
Image-to-Text • 401B • Updated • 5.73k • 42 -
unsloth/MiniMax-M2.1-GGUF
Text Generation • 229B • Updated • 148k • 148
Models that run well on a *standalone* RTX a6000's 48gb of VRAM.
The best little card under 100 euros. Full Precision vs Quants not benchmarked. This card is so much better at running inference than you realize.
-
unsloth/Nemotron-3-Nano-30B-A3B-GGUF
Text Generation • 32B • Updated • 102k • 229 -
unsloth/GLM-4.7-Flash-GGUF
Text Generation • 30B • Updated • 112k • 235 -
black-forest-labs/FLUX.2-dev
Image-to-Image • Updated • 106k • • 1.27k -
unsloth/Llama-4-Scout-17B-16E-Instruct-GGUF
Image-to-Text • 108B • Updated • 21k • 129
Toaster Tier but not iGPU
TPS can be as low as 1.0, seriously. its SLOW.
-
unsloth/GLM-4.7-GGUF
Text Generation • 358B • Updated • 130k • 179 -
unsloth/DeepSeek-R1-0528-GGUF
Text Generation • 671B • Updated • 4.44k • 193 -
unsloth/Llama-4-Maverick-17B-128E-Instruct-GGUF
Image-to-Text • 401B • Updated • 5.73k • 42 -
unsloth/MiniMax-M2.1-GGUF
Text Generation • 229B • Updated • 148k • 148
The stuff we actually use, pruned on an ongoing basis.