Zheng Han
traphix
ยท
AI & ML interests
None yet
Recent Activity
new activity
12 days ago
RedHatAI/DeepSeek-R1-0528-quantized.w4a16:Is 4 x H20 96G sufficient to run this model?
new activity
24 days ago
Qwen/Qwen3-30B-A3B-FP8:Remove vLLM FP8 Limitation
new activity
26 days ago
RedHatAI/Qwen3-235B-A22B-FP8-dynamic:Error running on A100?
Organizations
None yet
traphix's activity
Is 4 x H20 96G sufficient to run this model?
1
#2 opened 12 days ago
by
milongwong
Remove vLLM FP8 Limitation
10
#2 opened about 2 months ago
by
simon-mo
Error running on A100?
2
#4 opened about 1 month ago
by
traphix
Any plans for int8 quantized.w8a8?
#5 opened about 1 month ago
by
traphix
How about int8 quantization?
#3 opened about 1 month ago
by
traphix
How many RAM in GBs when quantizing Qwen3-235B-A22B?
#2 opened about 1 month ago
by
traphix
Where are the safetensors?
๐
1
1
#1 opened about 1 month ago
by
traphix
What is the difference between Qwen/Qwen3-32B-FP8 and this quatinized model๏ผ
4
#1 opened about 1 month ago
by
traphix
Does vllm 0.8.4 support this quantized model?
1
#1 opened about 2 months ago
by
traphix
This model beats Qwen Max!
๐
1
7
#33 opened 3 months ago
by
MrDevolver
why "MLA is not supported with awq_marlin quantization. Disabling MLA." with 4090 * 32 (4 node / vllm 0.7.2)
๐
1
3
#14 opened 4 months ago
by
FightLLM
Is there any accuracy results comparing to original DeepSeek-R1๏ผ
2
#15 opened 4 months ago
by
traphix
Has anyone evaluated the performance of the AWQ version of the model on benchmarks?
4
#8 opened 4 months ago
by
liuqianchao
skips the thinking process
11
#5 opened 4 months ago
by
muzizon
Deployment framework
27
#2 opened 5 months ago
by
xro7
vllm support a100
17
#2 opened 5 months ago
by
HuggingLianWang
Can it run on A100/A800 with VLLM?
3
#1 opened 11 months ago
by
Parkerlambert123