Alex semenov

LIOSAM007's activity

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

New activity in Comfy-Org/MiniMax-H3-Image about 20 hours ago

12 GB card numbers: int8 DiT + full 32B text encoder (RTX 4070 SUPER)

🔥 1

1

#12 opened about 20 hours ago by

LIOSAM007

New activity in PrunaAI/P-H3-Image-Comfy about 20 hours ago

P-H3-Image vs official H3-Image on a 12 GB card (int8, same prompts)

🔥 1

1

#12 opened about 20 hours ago by

LIOSAM007

New activity in Maximilian228/Qwen3.8-Flash-Next-NVFP4-GPTQ-Strata 6 days ago

Benchmark: RTX 4070 SUPER 12 GB + 96 GB DDR5 (Windows 11), 262K with --kv-resident

1

#2 opened 12 days ago by

LIOSAM007

New activity in bottlecapai/ThinkingCap-Qwen3.8-Flash-Next 6 days ago

Running ThinkingCap on the strata-nvfp4 fork (12 GB GPU) by swapping 372 tensors

❤️ 4

1

#2 opened 6 days ago by

LIOSAM007

New activity in LBH-123-AI/Minimax_h3_latent_Upscaler 7 days ago

Used as the second pass in a long-take chain

#25 opened 7 days ago by

LIOSAM007

New activity in MiniMaxAI/MiniMax-H3 7 days ago

One continuous 39-second shot from H3 segments, notes from a 12 GB card

#124 opened 7 days ago by

LIOSAM007

New activity in Comfy-Org/MiniMax-H3 7 days ago

39-second single takes on a 12 GB card with these files

#75 opened 7 days ago by

LIOSAM007

New activity in huihui-ai/Huihui-GLM-5.3-Flash-abliterated-GGUF 12 days ago

Maya v1.1.42 on 12 GB + 96 GB: 30K prompt stalls with --prefill auto

#8 opened 12 days ago by

LIOSAM007

New activity in orcarouter/Qwen3.8-Flash-Next-Uncensored-GGUF 12 days ago

IQ4_XS speed on 12 GB + 96 GB, and a note for Strata users

#20 opened 12 days ago by

LIOSAM007

New activity in AtomicChat/DeepSeek-V4.1-Flash-1017-Uncensored-GSQ-RCO-GGUF 12 days ago

Strata 0.2.8 on 12 GB + 96 GB: speed, CPU scaling, quality vs Flash-Next

👍 1

1

#1 opened 12 days ago by

LIOSAM007

New activity in SC117/Qwen3.8-Flash-Next-GSQ-RCO-abliterated-GGUF 12 days ago

IQ3_XXS speed on 12 GB + 96 GB: llama.cpp vs Strata

#9 opened 12 days ago by

LIOSAM007

New activity in Maximilian228/Qwen3.8-Flash-Next-NVFP4-GPTQ-Strata 12 days ago

Benchmark: RTX 4070 SUPER 12 GB + 96 GB DDR5 (Windows 11), 262K with --kv-resident

1

#2 opened 12 days ago by

LIOSAM007