Jonna Marie Matthiesen
Add accuracy tables, RTX 3500 Ada latency, and fix 270M variant grouping
7671b7c | model_family,model,type,batch,device,res,fps,frames,e2e,tps,tpot,ttft | |
| Qwen3-1.7B,Qwen/Qwen3-1.7B,text,1,agx_thor,N/A,N/A,N/A,5.1915,49.31,20.28,0.00 | |
| Qwen3-1.7B,embedl/Qwen3-1.7B-FlashHead,text,1,agx_thor,N/A,N/A,N/A,4.1107,62.28,16.06,0.00 | |
| Qwen3-1.7B,embedl/Qwen3-1.7B-FlashHead-W4A16,text,1,agx_thor,N/A,N/A,N/A,1.3543,189.02,5.29,0.00 | |
| Qwen3-1.7B,embedl/Qwen3-1.7B-W4A16,text,1,agx_thor,N/A,N/A,N/A,2.0541,124.63,8.02,0.00 | |
| Qwen3-0.6B,Qwen/Qwen3-0.6B,text,1,agx_thor,N/A,N/A,N/A,1.8237,140.38,7.12,0.00 | |
| Qwen3-0.6B,embedl/Qwen3-0.6B-FlashHead,text,1,agx_thor,N/A,N/A,N/A,1.4494,176.63,5.66,0.00 | |
| Qwen3-1.7B,Qwen/Qwen3-1.7B,text,1,agx_orin,N/A,N/A,N/A,5.5862,45.83,21.82,0.00 | |
| Qwen3-1.7B,embedl/Qwen3-1.7B-FlashHead,text,1,agx_orin,N/A,N/A,N/A,4.9568,51.65,19.36,0.00 | |
| Qwen3-1.7B,embedl/Qwen3-1.7B-FlashHead-W4A16,text,1,agx_orin,N/A,N/A,N/A,1.9020,134.59,7.43,0.00 | |
| Qwen3-1.7B,embedl/Qwen3-1.7B-W4A16,text,1,agx_orin,N/A,N/A,N/A,2.5357,100.96,9.91,0.00 | |
| Qwen3-0.6B,Qwen/Qwen3-0.6B,text,1,agx_orin,N/A,N/A,N/A,2.1783,117.52,8.51,0.00 | |
| Qwen3-0.6B,embedl/Qwen3-0.6B-FlashHead,text,1,agx_orin,N/A,N/A,N/A,1.8865,135.70,7.37,0.00 | |
| Qwen3-1.7B,Qwen/Qwen3-1.7B,text,1,rtx_3500_ada,N/A,N/A,N/A,,100.00,, | |
| Qwen3-1.7B,embedl/Qwen3-1.7B-FlashHead,text,1,rtx_3500_ada,N/A,N/A,N/A,,114.00,, | |
| Qwen3-1.7B,embedl/Qwen3-1.7B-W4A16,text,1,rtx_3500_ada,N/A,N/A,N/A,,206.00,, | |
| Qwen3-1.7B,embedl/Qwen3-1.7B-FlashHead-W4A16,text,1,rtx_3500_ada,N/A,N/A,N/A,,271.00,, | |