Tesla P40
41 records across 3 models (quants: EXL2 4.0bpw H6, Q2_K, Q3_K_M, Q4_0, Q4_K_M).
| hardware | model | quant | backend | tok/s | pp tok/s | W | ttft s | ctx/tg | date | class | flags | source |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Tesla P40 | CodeLlama-34B 34B | EXL2 4.0bpw H6 | ExLlamaV2 | 1.1905 | – | 80 | – | – | 2023-09 | community | – | source |
| Tesla P40 | Llama 3.1 8B | Q4_0 | llama.cpp (CUDA) | 54.75 | – | – | – | tg128 | 2026-10 | community | – | source |
| Tesla P40 | Llama 3.1 8B | Q4_0 | llama.cpp (CUDA) | 53.69 | – | – | – | 496 | 2026-10 | community | – | source |
| Tesla P40 | Llama 3.1 8B | Q4_0 | llama.cpp (CUDA) | 52.28 | – | – | – | tg128 | 2026-10 | community | – | source |
| Tesla P40 | Llama 3.1 8B | Q4_0 | llama.cpp (CUDA) | 51.87 | – | – | – | 2016 | 2026-10 | community | – | source |
| Tesla P40 | Llama 3.1 8B | Q4_0 | llama.cpp (CUDA) | 51.39 | – | – | – | 496 | 2026-10 | community | – | source |
| Tesla P40 | Llama 3.1 8B | Q4_0 | llama.cpp (CUDA) | 49.72 | – | – | – | 2016 | 2026-10 | community | – | source |
| Tesla P40 | Llama 3.1 8B | Q4_K_M | llama.cpp (CUDA) | 48.94 | – | – | – | tg128 | 2026-10 | community | – | source |
| Tesla P40 | Llama 3.1 8B | Q4_K_M | llama.cpp (CUDA) | 48.34 | – | – | – | tg128 | 2026-10 | community | – | source |
| Tesla P40 | Llama 3.1 8B | Q4_K_M | llama.cpp (CUDA) | 48.04 | – | – | – | 496 | 2026-10 | community | – | source |
| Tesla P40 | Llama 3.1 8B | Q4_K_M | llama.cpp (CUDA) | 47.47 | – | – | – | 496 | 2026-10 | community | – | source |
| Tesla P40 | Llama 3.1 8B | Q4_K_M | llama.cpp (CUDA) | 46.53 | – | – | – | 2016 | 2026-10 | community | – | source |
| Tesla P40 | Llama 3.1 8B | Q4_K_M | llama.cpp (CUDA) | 46.05 | – | – | – | 2016 | 2026-10 | community | – | source |
| Tesla P40 | Mixtral 8x7B | Q2_K | llama.cpp (CUDA) | 32.42 | – | – | – | tg128 | 2026-10 | community | – | source |
| Tesla P40 | Mixtral 8x7B | Q2_K | llama.cpp (CUDA) | 32.03 | – | – | – | 496 | 2026-10 | community | – | source |
| Tesla P40 | Mixtral 8x7B | Q2_K | llama.cpp (CUDA) | 31.37 | – | – | – | 2016 | 2026-10 | community | – | source |
| Tesla P40 | Mixtral 8x7B | Q2_K | llama.cpp (CUDA) | 30.65 | – | – | – | tg128 | 2026-10 | community | – | source |
| Tesla P40 | Mixtral 8x7B | Q2_K | llama.cpp (CUDA) | 30.30 | – | – | – | 496 | 2026-10 | community | – | source |
| Tesla P40 | Mixtral 8x7B | Q2_K | llama.cpp (CUDA) | 29.69 | – | – | – | 2016 | 2026-10 | community | – | source |
| Tesla P40 | Mixtral 8x7B | Q3_K_M | llama.cpp (CUDA) | 27.12 | – | – | – | tg128 | 2026-10 | community | – | source |
| Tesla P40 | Mixtral 8x7B | Q3_K_M | llama.cpp (CUDA) | 26.83 | – | – | – | 496 | 2026-10 | community | – | source |
| Tesla P40 | Mixtral 8x7B | Q3_K_M | llama.cpp (CUDA) | 26.35 | – | – | – | 2016 | 2026-10 | community | – | source |
| Tesla P40 | Mixtral 8x7B | Q3_K_M | llama.cpp (CUDA) | 25.76 | – | – | – | tg128 | 2026-10 | community | – | source |
| Tesla P40 | Mixtral 8x7B | Q3_K_M | llama.cpp (CUDA) | 25.54 | – | – | – | 496 | 2026-10 | community | – | source |
| Tesla P40 | Mixtral 8x7B | Q3_K_M | llama.cpp (CUDA) | 25.06 | – | – | – | 2016 | 2026-10 | community | – | source |
| Tesla P40 | Mixtral 8x7B | Q2_K | llama.cpp (CUDA) | – | 33.71 | – | – | – | 2026-10 | community | – | source |
| Tesla P40 | Mixtral 8x7B | Q2_K | llama.cpp (CUDA) | – | 31.82 | – | – | – | 2026-10 | community | – | source |
| Tesla P40 | Mixtral 8x7B | Q2_K | llama.cpp (CUDA) | – | 46.65 | – | – | – | 2026-10 | community | – | source |
| Tesla P40 | Mixtral 8x7B | Q2_K | llama.cpp (CUDA) | – | 45.35 | – | – | – | 2026-10 | community | – | source |
| Tesla P40 | Mixtral 8x7B | Q2_K | llama.cpp (CUDA) | – | 51.35 | – | – | – | 2026-10 | community | – | source |
| Tesla P40 | Mixtral 8x7B | Q2_K | llama.cpp (CUDA) | – | 50.64 | – | – | – | 2026-10 | community | – | source |
| Tesla P40 | Mixtral 8x7B | Q2_K | llama.cpp (CUDA) | – | 54.41 | – | – | – | 2026-10 | community | – | source |
| Tesla P40 | Mixtral 8x7B | Q2_K | llama.cpp (CUDA) | – | 54.33 | – | – | – | 2026-10 | community | – | source |
| Tesla P40 | Mixtral 8x7B | Q3_K_M | llama.cpp (CUDA) | – | 26.74 | – | – | – | 2026-10 | community | – | source |
| Tesla P40 | Mixtral 8x7B | Q3_K_M | llama.cpp (CUDA) | – | 28.14 | – | – | – | 2026-10 | community | – | source |
| Tesla P40 | Mixtral 8x7B | Q3_K_M | llama.cpp (CUDA) | – | 32.71 | – | – | – | 2026-10 | community | – | source |
| Tesla P40 | Mixtral 8x7B | Q3_K_M | llama.cpp (CUDA) | – | 36.54 | – | – | – | 2026-10 | community | – | source |
| Tesla P40 | Mixtral 8x7B | Q3_K_M | llama.cpp (CUDA) | – | 35.99 | – | – | – | 2026-10 | community | – | source |
| Tesla P40 | Mixtral 8x7B | Q3_K_M | llama.cpp (CUDA) | – | 40.58 | – | – | – | 2026-10 | community | – | source |
| Tesla P40 | Mixtral 8x7B | Q3_K_M | llama.cpp (CUDA) | – | 57.55 | – | – | – | 2026-10 | community | – | source |
| Tesla P40 | Mixtral 8x7B | Q3_K_M | llama.cpp (CUDA) | – | 57.53 | – | – | – | 2026-10 | community | – | source |
All rows cite their source. Provenance: community 40018 sourced 1085