nm-testing/SparseLlama-3.1-8B-gsm8k-pruned.2of4-chnl_wts_tensor_act_fp8-BitM
5B • Updated • 18
nm-testing/TinyLlama-1.1B-Chat-v1.0-gsm8k-pruned.2of4-tensor_wts_tensor_act_int8-BitM
0.7B • Updated • 38
nm-testing/SparseLlama-3.1-8B-gsm8k-pruned.2of4-chnl_wts_per_tok_dyn_act_int8-BitM
5B • Updated • 16
nm-testing/TinyLlama-1.1B-Chat-v1.0-gsm8k-pruned.2of4-tensor_wts_tensor_act_fp8-BitM
0.7B • Updated • 40
nm-testing/TinyLlama-1.1B-Chat-v1.0-gsm8k-pruned.2of4-tensor_wts_per_tok_dyn_act_int8-BitM
0.7B • Updated • 38
nm-testing/SparseLlama-3.1-8B-gsm8k-pruned.2of4-chnl_wts_per_tok_dyn_act_fp8-BitM
5B • Updated • 18
nm-testing/TinyLlama-1.1B-Chat-v1.0-gsm8k-pruned.2of4-tensor_wts_per_tok_dyn_act_fp8-BitM
0.7B • Updated • 40
nm-testing/TinyLlama-1.1B-Chat-v1.0-gsm8k-pruned.2of4-chnl_wts_tensor_act_int8-BitM
0.7B • Updated • 34
nm-testing/TinyLlama-1.1B-Chat-v1.0-gsm8k-pruned.2of4-chnl_wts_tensor_act_fp8-BitM
0.7B • Updated • 40
nm-testing/TinyLlama-1.1B-Chat-v1.0-gsm8k-pruned.2of4-chnl_wts_per_tok_dyn_act_fp8-BitM
0.7B • Updated • 40
nm-testing/TinyLlama-1.1B-Chat-v1.0-gsm8k-pruned.2of4-chnl_wts_per_tok_dyn_act_int8-BitM
0.7B • Updated • 37
nm-testing/TinyLlama-1.1B-Chat-v1.0-gsm8k-pruned.2of4-Uncompressed
1B • Updated • 6
nm-testing/pixtral-12b-w4a16-actorder-weight
13B • Updated • 15
nm-testing/pixtral-12b-w8a8-noactorder
13B • Updated • 18
nm-testing/pixtral-12b-w4a16-actorder-group
13B • Updated • 19
nm-testing/TinyLlama-1.1B-Chat-v1.0-pruned.2of4-Sparse24BitMaskCompressed
0.7B • Updated • 12
nm-testing/TinyLlama-1.1B-Chat-v1.0-pruned.2of4-Sparse24ByteMaskCompressed
2B • Updated • 12
nm-testing/TinyLlama-1.1B-Chat-v1.0-sparse24-BitMaskCompressed
2B • Updated • 15
nm-testing/Meta-Llama-3-8B-Instruct-W8A8-Static-Per-Tensor-Asym
8B • Updated • 46
nm-testing/Llama-3.2-1B-Instruct-W8A8-Static-Per-Tensor-Asym
1B • Updated • 18
nm-testing/TinyLlama-1.1B-Chat-v1.0-pruned_50.0-2of4-BitMaskCompressed
0.7B • Updated • 32
nm-testing/llama2.c-stories15M-pruned_50.0-2of4-BitMaskCompressed
13M • Updated • 65
nm-testing/Meta-Llama-3-8B-Instruct-W8A8-Dynamic-Asym
8B • Updated • 122
nm-testing/Meta-Llama-3-8B-Instruct-W8A8-Static-Per-Tensor-Sym
8B • Updated • 111
nm-testing/SparseLlama-3.1-8B-gsm8k-pruned.2of4-channel_weights_per_token_dynamic_act_fp8-BitMaskCompressed
1B • Updated • 35
nm-testing/SparseLlama-3.1-8B-gsm8k-pruned.2of4-tensor_weights_tensor_act_fp8-BitMaskCompressed
5B • Updated • 35
nm-testing/SparseLlama-3.1-8B-gsm8k-pruned.2of4-channel_weights_tensor_act_fp8-BitMaskCompressed
5B • Updated • 35
nm-testing/SparseLlama-3.1-8B-gsm8k-pruned.2of4-tensor_weights_per_token_dynamic_act_fp8-BitMaskCompressed
5B • Updated • 36
nm-testing/llama2.c-stories15M-pruned_50.2of4-uncompressed
24.4M • Updated • 52
nm-testing/SparseLlama-3.1-8B-gsm8k-pruned.2of4-W8A8-testing
8B • Updated • 22