dealign.ai PRO

dealignai's activity

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

New activity in IFM/K2-Horizon-7B 3 days ago

I wish you delivered what you say you delivered

👍👀 4

15

#3 opened about 1 month ago by

NezTheNaughty

New activity in dealignai/Bonsai-2-27B-Ternary-CRACK-GGUF 23 days ago

reasoning mode still looping

8

#1 opened 23 days ago by

Jcamacho05

New activity in dealignai/DeepSeek-V4.1-Flash-UNCENSORED-FP8 30 days ago

Field report: UNCENSORED-FP8 serving at 1M context on 8x DGX Spark (GB10), vLLM TP8, with numbers

🔥 2

6

#1 opened about 1 month ago by

0xMagnus

New activity in dealignai/GLM-5.3-CYBERSECURITY-FP8 about 1 month ago

NVFP4

➕ 2

2

#2 opened about 1 month ago by

alonsoko

New activity in dealignai/GLM-5.3-UNCENSORED-FP8 about 1 month ago

Field report: FP8 v2 and NVFP4 on 8x DGX Spark (GB10) at TP=8 -- what works, what bites, and one template trap worth documenting

8

#3 opened about 1 month ago by

0xMagnus

New activity in dealignai/Qwen3.8-Flash-Next-UNCENSORED-FP8 about 1 month ago

Reasoning-trace length: abliterated vs. base Qwen3.8-Flash-Next on GPQA Diamond

13

#2 opened about 1 month ago by

Soomin33

New activity in dealignai/Qwen3.8-Flash-Next-CYBERSECURITY-NVFP4 about 1 month ago

other formats

9

#2 opened about 1 month ago by

oneerrorx

New activity in dealignai/GLM-5.3-UNCENSORED-FP8 about 1 month ago

V2 still degenerates into word salad on real long-context agent workloads

2

#2 opened about 1 month ago by

leonice2

New activity in dealignai/Qwen3.8-Flash-Next-CYBERSECURITY-NVFP4 about 1 month ago

Reasoning improvements?

1

#1 opened about 1 month ago by

dealignai

New activity in dealignai/GLM-5.3-Flash-UNCENSORED-NVFP4 about 1 month ago

Token output random corruption, ModelOpt artifact?

1

#2 opened about 1 month ago by

m-e-acc

Vision not working

3

#1 opened about 1 month ago by

GulfCoastAI

New activity in dealignai/GLM-5.3-Flash-UNCENSORED-FP8 about 1 month ago

GGUF?

5

#1 opened about 2 months ago by

kabachuha

New activity in dealignai/Qwen3.8-27B-UNCENSORED-GGUF about 2 months ago

ty dealignai <3

👍 2

4

#2 opened about 2 months ago by

0xHgageface

Feature request — CRACK + Claude Opus 4.7-style reasoning distillation

1

#1 opened about 2 months ago by

imedox

New activity in stamsam/Gemma-4-21B-REAP-JANG3M-MLX 4 months ago

Interesting work!

1

#1 opened 4 months ago by

dealignai

New activity in dealignai/Gemma-4-31B-JANG_4M-CRACK 5 months ago

Fix loading in LM Studio / mlx-llm: quant map keys + vision weights

4

#25 opened 5 months ago by

IsaacLiu

New activity in dealignai/Gemma-4-26B-A4B-JANG_4M-CRACK 5 months ago

Gemma 4 26B (A4B) - Garbage output (korean/chinese) on MLX 0.31.2 while 31B works fine

3

#2 opened 6 months ago by

Z3NN001

New activity in OsaurusAI/MiniMax-M2.7-JANGTQ_K 5 months ago

Add MMLU-200 (93.5%), speed/memory benchmarks, fix variants table sizes (47→56 GB), expand topic tags

#1 opened 5 months ago by

dealignai

New activity in dealignai/Gemma-4-31B-JANG_4M-CRACK 6 months ago

gemma4-crack

#8 opened 6 months ago by

Adam666Eves

Bug

3

#3 opened 6 months ago by

Zayaan