Alex Cardo

alexcardo's activity

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

New activity in cyankiwi/gemma-4-31B-it-AWQ-8bit 1 day ago

Quant HAS issues + results with vLLM on 8x 3090

4

#1 opened 9 days ago by

dehnhaide

New activity in RedHatAI/gemma-4-31B-it-NVFP4 1 day ago

Please update tokenizer config as well

12

#2 opened 2 days ago by

alexcardo

New activity in RedHatAI/gemma-4-31B-it-NVFP4 2 days ago

Can you plese update the chat template

❤️ 3

1

#1 opened 2 days ago by

alexcardo

New activity in LilaRest/gemma-4-31B-it-NVFP4-turbo 3 days ago

Is this quant support image recognition?

6

#1 opened 3 days ago by

alexcardo

New activity in google/gemma-4-31B-it 3 days ago

Chat template is too complicated that even Gemma 4 itself has no idea how to parse it

1

#53 opened 3 days ago by

alexcardo

New activity in nvidia/Gemma-4-31B-IT-NVFP4 6 days ago

Why is this 4bit version has a 32.7 GB size?

➕ 3

17

#3 opened 10 days ago by

alexcardo

New activity in google/gemma-4-31B-it 7 days ago

Why Gemma4 can't recognize the entire text on image?

🚀 4

6

#12 opened 9 days ago by

alexcardo

New activity in nvidia/Gemma-4-31B-IT-NVFP4 9 days ago

这个版本对于5090单卡来说还是太大了

10

#4 opened 10 days ago by

iwaitu

New activity in Qwen/Qwen3.5-27B-GPTQ-Int4 21 days ago

30.3 GB?

👀 4

3

#6 opened 25 days ago by

pedalnomica

updated a Space 26 days ago

Sleeping

Startpage

📈

published a Space 26 days ago

Sleeping

Startpage

📈

New activity in osoleve/Qwen3.5-27B-Text-NVFP4-MTP about 1 month ago

Shitty results compared to regular NVFP4 without MTP

4

#3 opened about 1 month ago by

alexcardo

New activity in Qwen/Qwen3.5-27B-GPTQ-Int4 about 1 month ago

怎么和fp8一样大

👍👀 21

2

#1 opened about 1 month ago by

chenzin23

New activity in Qwen/Qwen3.5-27B about 1 month ago

Russian language support, bad grammar!

13

#12 opened about 2 months ago by

alexcardo

New activity in Sangto/Seed-X-PPO-7B-Q8_0-GGUF 9 months ago

It doesn't translate even the example from the system prompt

1

#1 opened 9 months ago by

alexcardo

New activity in deepseek-ai/DeepSeek-R1-0528 11 months ago

When will it be available via API

7

#27 opened 11 months ago by

alexcardo

New activity in Qwen/Qwen3-32B 12 months ago

After setting /nothinking or enable_thinking=False, can the empty <thinking> tag be omitted from the response?

👍 4

2

#13 opened 12 months ago by

pteromyini

New activity in lmstudio-community/gemma-3-12B-it-qat-GGUF 12 months ago

No difference in size ?

2

#2 opened 12 months ago by

Pumba2

New activity in SuperAnnotate/ai-detector about 1 year ago

What does the final number mean?

1

#3 opened about 1 year ago by

alexcardo

New activity in lmstudio-community/Phi-4-mini-instruct-GGUF about 1 year ago

unknown pre-tokenizer type: 'gpt-4o'

5

#1 opened about 1 year ago by

alexcardo