A6000
20 records across 2 models (quants: EXL2 4bit, GPTQ 4bit, Q4_0).
| hardware | model | quant | backend | tok/s | pp tok/s | W | ttft s | ctx/tg | date | class | flags | source |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| A6000 | FreeWilly2 70B | EXL2 4bit | ExLlamaV2 | 10.97 | – | – | – | 1046 | 2023-09 | community | – | source |
| A6000 | FreeWilly2 70B | EXL2 4bit | ExLlamaV2 | 10.42 | – | – | – | 1404 | 2023-09 | community | – | source |
| A6000 | FreeWilly2 70B | EXL2 4bit | ExLlamaV2 | 9.96 | – | – | – | 941 | 2023-09 | community | – | source |
| A6000 | FreeWilly2 70B | EXL2 4bit | ExLlamaV2 | 9.73 | – | – | – | 1880 | 2023-09 | community | – | source |
| A6000 | FreeWilly2 70B | EXL2 4bit | ExLlamaV2 | 9.64 | – | – | – | 941 | 2023-09 | community | – | source |
| A6000 | FreeWilly2 70B | EXL2 4bit | ExLlamaV2 | 9.38 | – | – | – | 941 | 2023-09 | community | – | source |
| A6000 | FreeWilly2 70B | EXL2 4bit | ExLlamaV2 | 9.16 | – | – | – | 2370 | 2023-09 | community | – | source |
| A6000 | FreeWilly2 70B | GPTQ 4bit | ExLlamaV2 | 8.81 | – | – | – | 1040 | 2023-09 | community | – | source |
| A6000 | FreeWilly2 70B | GPTQ 4bit | ExLlamaV2 | 8.45 | – | – | – | 1483 | 2023-09 | community | – | source |
| A6000 | FreeWilly2 70B | EXL2 4bit | ExLlamaV2 | 8.43 | – | – | – | 3027 | 2023-09 | community | – | source |
| A6000 | FreeWilly2 70B | GPTQ 4bit | ExLlamaV2 | 8.07 | – | – | – | 1967 | 2023-09 | community | – | source |
| A6000 | FreeWilly2 70B | GPTQ 4bit | ExLlamaV2 | 7.91 | – | – | – | 941 | 2023-09 | community | – | source |
| A6000 | FreeWilly2 70B | GPTQ 4bit | ExLlamaV2 | 7.78 | – | – | – | 2509 | 2023-09 | community | – | source |
| A6000 | FreeWilly2 70B | GPTQ 4bit | ExLlamaV2 | 7.56 | – | – | – | 2860 | 2023-09 | community | – | source |
| A6000 | FreeWilly2 70B | GPTQ 4bit | ExLlamaV2 | 7.52 | – | – | – | 941 | 2023-09 | community | – | source |
| A6000 | FreeWilly2 70B | EXL2 4bit | ExLlamaV2 | 7.49 | – | – | – | 3551 | 2023-09 | community | – | source |
| A6000 | FreeWilly2 70B | GPTQ 4bit | ExLlamaV2 | 7.38 | – | – | – | 3182 | 2023-09 | community | – | source |
| A6000 | FreeWilly2 70B | GPTQ 4bit | ExLlamaV2 | 7.07 | – | – | – | 3588 | 2023-09 | community | – | source |
| A6000 | llama 7B | Q4_0 | llama.cpp (CUDA) | 144.87 | 5662.39 | – | – | tg128 | 2025-09-01 | community | – | source |
| A6000 | llama 7B | Q4_0 | llama.cpp (CUDA) | 138.73 | 4913.93 | – | – | tg128 | 2025-09-01 | community | – | source |
All rows cite their source. Provenance: community 40018 sourced 1085