Hugging Face's logo
Phil's picture
39 1 40

Phil

phil111
Gargaz's profile picture Mi6paulino's profile picture altomek's profile picture
·

AI & ML interests

None yet

Recent Activity

liked a model 1 day ago
google/gemma-3-27b-it
new activity 4 days ago
Qwen/Qwen3-30B-A3B:Qwen3 is great, but could be better.
new activity 8 days ago
Qwen/Qwen3-235B-A22B:Qwen is loosing broad knowledge since Qwen2.
View all activity

Organizations

None yet

phil111's activity

liked a model 1 day ago
New activity in Qwen/Qwen3-30B-A3B 4 days ago

Qwen3 is great, but could be better.

7
19
#18 opened 6 days ago by
phil111
New activity in Qwen/Qwen3-235B-A22B 8 days ago

Qwen is loosing broad knowledge since Qwen2.

9
14
#16 opened 9 days ago by
phil111
New activity in Qwen/Qwen3-30B-A3B 9 days ago

Very fast and powerful, but with one glaring weakness.

1
2
#6 opened 9 days ago by
phil111
New activity in THUDM/GLM-4-32B-0414 22 days ago

SimpleQA Scores Are WAY off

4
5
#3 opened 23 days ago by
phil111
New activity in meta-llama/Llama-4-Scout-17B-16E-Instruct 25 days ago

Less Knowledge Than Llama 3.3 70b?

2
5
#60 opened 27 days ago by
phil111
New activity in meta-llama/Llama-4-Scout-17B-16E-Instruct 29 days ago

13 B and34 B Pleeease!!! Most people cannot even run this.

4
4
#52 opened 29 days ago by
UniversalLove333
New activity in deepseek-ai/DeepSeek-V3-0324 about 1 month ago

SimpleQA?

7
3
#29 opened about 1 month ago by
phil111
New activity in mistralai/Mistral-Small-24B-Instruct-2501 3 months ago

This Mistral Small has FAR less knowledge than the last.

5
20
#5 opened 3 months ago by
phil111
liked a model 4 months ago
New activity in internlm/internlm3-8b-instruct 4 months ago

English tests and tasks are absurdly overfit.

7
21
#8 opened 4 months ago by
phil111
New activity in microsoft/phi-4 4 months ago

A heavily filtered corpus simply doesn't work.

10
4
#19 opened 4 months ago by
phil111

I Don't Understand This Model

7
16
#9 opened 4 months ago by
phil111
New activity in matteogeniaccio/phi-4 4 months ago

Notably better than Phi3.5 in many ways, but something is wrong.

1
8
#5 opened 5 months ago by
phil111
liked a model 4 months ago
New activity in deepseek-ai/DeepSeek-V3-Base 4 months ago

Very impressive. Good world knowledge (SimpleQA of 25) despite high math/coding performance.

3
2
#27 opened 4 months ago by
phil111
liked a model 4 months ago
New activity in matteogeniaccio/phi-4 5 months ago

Notably better than Phi3.5 in many ways, but something is wrong.

1
8
#5 opened 5 months ago by
phil111
New activity in NyxKrage/Microsoft_Phi-4 5 months ago

SimpleQA score

2
#1 opened 5 months ago by
frappuccino
New activity in ibm-granite/granite-3.1-8b-instruct 5 months ago

Exceptional creative writer

15
5
#1 opened 5 months ago by
SubtleOne