No machine on this network has measured this model yet. Everything below is what the catalog knows exists — where the weights come from, how large each build is, and whose work it is. Nothing here is a claim about what the model can do; that only appears once a machine has run it and signed the result.
DevQuasar/baidu.ERNIE-4.5-300B-A47B-PT-GGUF:IQ1_Mllamacpp · IQ1_M · 62.5 GiB · quantizer (DevQuasar)
DevQuasar/baidu.ERNIE-4.5-300B-A47B-PT-GGUF:IQ2_XXSllamacpp · IQ2_XXS · 72.9 GiB · quantizer (DevQuasar)
baidu/ERNIE-4.5-300B-A47B-2Bits-TP2-Paddlevllm · BF16 · 87.7 GiB · vendor (baidu)
DevQuasar/baidu.ERNIE-4.5-300B-A47B-PT-GGUF:Q3_K_Mllamacpp · Q3_K_M · 132.8 GiB · quantizer (DevQuasar)
mlx-community/ERNIE-4.5-300B-A47B-PT-4bitmlx · 4BIT · 156.9 GiB · quantizer (mlx-community)
DevQuasar/baidu.ERNIE-4.5-300B-A47B-PT-GGUF:Q4_K_Mllamacpp · Q4_K_M · 167.8 GiB · quantizer (DevQuasar)
DevQuasar/baidu.ERNIE-4.5-300B-A47B-PT-GGUF:Q5_K_Mllamacpp · Q5_K_M · 197.4 GiB · quantizer (DevQuasar)
DevQuasar/baidu.ERNIE-4.5-300B-A47B-PT-GGUF:Q8_0llamacpp · Q8_0 · 296.4 GiB · quantizer (DevQuasar)
DevQuasar/baidu.ERNIE-4.5-300B-A47B-PT-GGUF:F16llamacpp · F16 · 557.9 GiB · quantizer (DevQuasar)