| license: apache-2.0 | |
| base_model: | |
| - cognitivecomputations/dolphin-2.7-mixtral-8x7b | |
| GGUF IQ3_M quant of cognitivecomputations/dolphin-2.7-mixtral-8x7b (both non-imatrix and imatrix) | |
| It fits into 24GiB VRAM with 32768 context (@ 8bit KV cache quantization). |