Aeon Forge's picture

Aeon Forge

AEON-7

·

AI & ML interests

Systems engineer. Security researcher. Quiet architect of things that probably shouldn't work — but do. Small models doing big things. Quantization, fine-tuning, skill training, agentic workflows, and generative CAD. Been at neural nets since 2013, still surprised weekly. The GPU hums, the coffee's on, and the door's open.

Recent Activity

updated a model about 10 hours ago

AEON-7/Gemma-4-26B-A4B-it-Uncensored-NVFP4

new activity about 22 hours ago

AEON-7/Qwen3.6-27B-AEON-Ultimate-Uncensored-NVFP4:Using RTX 5090

new activity about 22 hours ago

AEON-7/Qwen3.6-27B-AEON-Ultimate-Uncensored-NVFP4:[Request] modelopt NVFP4 + MTP variant with proper SSM layer preservation

View all activity

Organizations

None yet

New activity in AEON-7/Qwen3.6-27B-AEON-Ultimate-Uncensored-NVFP4 about 22 hours ago

Using RTX 5090

#3 opened 11 days ago by

[Request] modelopt NVFP4 + MTP variant with proper SSM layer preservation

#4 opened 10 days ago by

New activity in AEON-7/Qwen3.6-35B-A3B-heretic-NVFP4 about 23 hours ago

Image crash

#1 opened 7 days ago by

New activity in AEON-7/Qwen3.6-27B-AEON-Ultimate-Uncensored-NVFP4 about 23 hours ago

The DGX Spark didn't reach the benchmark speed.

#5 opened 10 days ago by

New activity in AEON-7/Qwen3.6-27B-AEON-Ultimate-Uncensored-BF16 5 days ago

Is MTP possible?

#2 opened 11 days ago by

MTP Grafted back in, findings...

#6 opened 8 days ago by

Spliting .safetensors files into smaller pieces?

#7 opened 6 days ago by

New activity in AEON-7/Qwen3.6-27B-AEON-Ultimate-Uncensored-BF16 7 days ago

[Request] Pre-Tokenizer Info

#5 opened 9 days ago by

ZuulTheMerovingian

New activity in AEON-7/Qwen3.6-27B-AEON-Ultimate-Uncensored-BF16 9 days ago

Add a `quant_config` block (or rename the repo) so bf16 vs NVFP4 isn't ambiguous

#4 opened 9 days ago by

New activity in AEON-7/Qwen3.6-27B-AEON-Ultimate-Uncensored-NVFP4 9 days ago

Speed is not fast

#2 opened 11 days ago by

New activity in AEON-7/Qwen3.6-27B-AEON-Ultimate-Uncensored-BF16 10 days ago

gguf please

#1 opened 12 days ago by

The BPE pre-tokenizer was not recognized!

#3 opened 10 days ago by

New activity in AEON-7/Gemma-4-31B-it-DECKARD-HERETIC-Uncensored-NVFP4 10 days ago

KeyError: 'layers.0.mlp.down_proj.pre_quant_scale'

#2 opened 10 days ago by

New activity in AEON-7/Qwen3.6-27B-AEON-Ultimate-Uncensored-NVFP4 12 days ago

Freezing during autotune

#1 opened 12 days ago by

New activity in AEON-7/DFlash-Qwen3.5-27B-Uncensored-NVFP4 14 days ago

ValueError: ModelOpt only supports specific quantizations in vLLM (also on nightly)

#2 opened 14 days ago by

New activity in AEON-7/supergemma4-26b-abliterated-multimodal-nvfp4 17 days ago

Scaling with concurrency?

#1 opened 20 days ago by

New activity in AEON-7/Gemma-4-31B-it-DECKARD-HERETIC-Uncensored-NVFP4 17 days ago

GGUF version?

#1 opened 17 days ago by

New activity in AEON-7/supergemma4-26b-abliterated-multimodal-nvfp4 19 days ago

Scaling with concurrency?

#1 opened 20 days ago by