Pirate Femme PRO

Babsie's activity

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

updated a model about 1 hour ago

Babsie/Maenad-70B

Text Generation • 71B • Updated about 1 hour ago

published a model about 5 hours ago

Babsie/Maenad-70B

Text Generation • 71B • Updated about 1 hour ago

updated 2 collections about 7 hours ago

70B and over possibilities

Collection

21 items • Updated about 7 hours ago

24B models

Collection

20 items • Updated about 7 hours ago

updated a collection about 9 hours ago

70B and over possibilities

Collection

21 items • Updated about 7 hours ago

updated a model about 10 hours ago

Babsie/NousResearch-Hermes-2-Theta-Llama-3-70B

Text Generation • 71B • Updated about 10 hours ago • 60

reacted to mitkox's post with 👀 about 10 hours ago

view post

Post

790

I just threw Qwen3-0.6B in BF16 into an on device AI drag race on AMD Strix Halo with vLLM:

564 tokens/sec on short 100-token sprints
96 tokens/sec on 8K-token marathons

TL;DR You don't just run AI on AMD. You negotiate with it.

The hardware absolutely delivers. Spoiler alert; there is exactly ONE configuration where vLLM + ROCm + Triton + PyTorch + Drivers + Ubuntu Kernel to work at the same time. Finding it required the patience of a saint

Consumer AMD for AI inference is the ultimate "budget warrior" play, insane performance-per-euro, but you need hardcore technical skills that would make a senior sysadmin nod in quiet respect.

See translation

• 

🔥

2

2

👀

1

1

+

Reply

updated a model about 11 hours ago

Babsie/Trickster-Theta-4-70B

71B • Updated about 11 hours ago • 129 • 2

updated a collection 1 day ago

24B and 34B models

Collection

1.Repairing/updating, merging to get Goblin: theatrical, anarchic, trickster-librarian assistant. 2. A smaller Trickster model. • 14 items • Updated about 1 hour ago

updated a collection 2 days ago

Mischievous Rake

Collection

Models I am repairing and/or adjusting and testing for specific trickster builds. Included are failures and broken bits. All labeled. • 14 items • Updated 2 days ago

updated 2 models 2 days ago

Babsie/unnamed-intermediate-merge-model-testA

Text Generation • 71B • Updated 2 days ago • 3

Babsie/TrckstrExtDimenBrainDamageNO_USE_RefernceOnly

Text Generation • 71B • Updated 2 days ago

New activity in DontPlanToEnd/UGI-Leaderboard 2 days ago

Eval Request

2

#414 opened 8 days ago by

Babsie

published a model 2 days ago

Babsie/TrckstrExtDimenBrainDamageNO_USE_RefernceOnly

Text Generation • 71B • Updated 2 days ago

liked 4 models 2 days ago

Vortex5/ChaosRose-24B

Text Generation • 24B • Updated May 29 • 5 • 2

NousResearch/DeepHermes-3-Mistral-24B-Preview

Text Generation • 24B • Updated Mar 13 • 1.67k • 119

NousResearch/Hermes-4-405B

Text Generation • 406B • Updated Sep 2 • 808 • • 73

Daemontatox/Zeril-Hera

Text Generation • 71B • Updated Sep 8 • 13 • 1

updated a collection 2 days ago

Mischievous Rake

Collection

Models I am repairing and/or adjusting and testing for specific trickster builds. Included are failures and broken bits. All labeled. • 14 items • Updated 2 days ago

updated a model 2 days ago

Babsie/Sao10K-Euryale-v2.1-70B-labmod

71B • Updated 2 days ago • 21