Timothy Rohrbaugh PRO

trohrbaugh's activity

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

New activity in trohrbaugh/Qwen3.8-Flash-Next-heretic-2 4 days ago

Please share the lora for trohrbaugh/Qwen3.8-Flash-Next-heretic-2

1

#1 opened 5 days ago by

MJHAU

New activity in Qwen/Qwen3.8-27B 8 days ago

Qwen 3.8 get stuck in massive 20k–50k token thinking loops on SGLang?

6

#178 opened 28 days ago by

undeadindustries

New activity in deepseek-ai/DeepSeek-V4.1-Flash 13 days ago

DeepSeek V4.1 Flash Lite

➕👍 36

16

#11 opened 14 days ago by

KeinNiemand

New activity in OBLITERATUS/Qwen3.8-27B-OBLITERATED 24 days ago

`chat_template.jinja` differs from the base — plain API calls return empty `content`, and `reasoning_effort` is inert

#14 opened 24 days ago by

trohrbaugh

New activity in trohrbaugh/Qwen3.8-27B-heretic-ara 24 days ago

LORAs at lower KLD ?

1

#5 opened 24 days ago by

DavidAU

New activity in trohrbaugh/Hy3-heretic 26 days ago

Hy3-FP8?

1

#2 opened 28 days ago by

AgileTurnip

New activity in trohrbaugh/Qwen3.8-27B-heretic-ara about 1 month ago

Will it support dflash2?

1

#4 opened about 1 month ago by

wanghf

NVFP4 Version

🔥 3

1

#3 opened about 1 month ago by

PassingByPixels

This model is still fairly prone to refusals in thinking mode

2

#1 opened about 1 month ago by

cchickk

https://github.com/timrohrbaugh/heretic gives 404

1

#2 opened about 1 month ago by

trithemius

New activity in deepseek-ai/DeepSeek-V4-Flash-0731 about 2 months ago

How can it be served with 4 x RTX Pro6000?

👀 1

19

#22 opened about 2 months ago by

niuchuang

New activity in RedHatAI/Inkling-Small-FP8-dynamic about 2 months ago

This quant needs fixing

2

#1 opened about 2 months ago by

trohrbaugh

New activity in trohrbaugh/Hy3-heretic 2 months ago

Can you create awq int 4 version?

1

#1 opened 2 months ago by

denisvasi

New activity in InternScience/Agents-A1 3 months ago

tool calling fail is an artifact of post training

#13 opened 3 months ago by

trohrbaugh

New activity in trohrbaugh/ZAYA1-74B-preview-heretic 5 months ago

IMO after trying it in VLLM with custom kernels

1

#1 opened 5 months ago by

coughmedicine

New activity in Zyphra/ZAYA1-74B-preview 5 months ago

Multi-GPU inference requires 6 patches to modeling_zaya.py

2

#2 opened 5 months ago by

trohrbaugh

New activity in trohrbaugh/Qwen3.6-27B-heretic-ara 5 months ago

Bug found

1

#2 opened 5 months ago by

trohrbaugh

could you make an uncensored version of DeepSeek v4 Flash?

2

#1 opened 5 months ago by

CCSSNE

New activity in trohrbaugh/Qwen3.6-27B-heretic 5 months ago

Quantized model available?

1

#1 opened 5 months ago by

dieselheart01

New activity in trohrbaugh/Qwen3.6-35B-A3B-heretic 5 months ago

missing tensor

3

#2 opened 5 months ago by

coughmedicine