Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
LLM Training Performance on Qwen models on A6000 (48GB VRAM) 2.5 (train)
Loading...
56.82
TFLOPS
MegaTrain
22.5728
31.4639
40.355
49.2461
Apr 6, 2026
TFLOPS
GPU Memory Usage (GB)
CPU Memory Usage (GB)
Throughput (tokens/s)
Updated 1mo ago
Evaluation Results
Method
Method
Links
TFLOPS
GPU Memory Usage (GB)
CPU Memory Usage (GB)
Throughput (tokens/s)
MegaTrain
Model=Qwen2.5-14B, Max...
2026.04
56.82
44.64
104.1
641
MegaTrain
Model=Qwen2.5-7B, Max...
2026.04
55.73
44.74
56.7
1,219
MegaTrain
Model=Qwen2.5-3B, Max...
2026.04
49.7
46.74
38.3
2,153
ZeRO-3 Offload
Model=Qwen2.5-7B, Max...
2026.04
27.55
20.83
-
-
ZeRO-3 Offload
Model=Qwen2.5-3B, Max...
2026.04
23.89
20.33
-
-
Feedback
Search any
task
Search any
task