cohere2moe 30B.A3B
8 records across 1 hardware.
| hardware | model | quant | backend | tok/s | pp tok/s | W | ttft s | ctx/tg | date | class | flags | source |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Arc Pro B60 | cohere2moe 30B.A3B | Q4_K - Medium | llama.cpp | 70.86 | 1048.0 | – | – | tg128 | 2026-08-29 | community | – | source |
| Arc Pro B60 | cohere2moe 30B.A3B | Q4_K - Medium | llama.cpp | 70.59 | 795.24 | – | – | tg128 | 2026-08-29 | community | – | source |
| Arc Pro B60 | cohere2moe 30B.A3B | Q4_K - Medium | llama.cpp | 67.51 | 779.19 | – | – | tg128 | 2026-08-29 | community | – | source |
| Arc Pro B60 | cohere2moe 30B.A3B | Q4_K - Medium | llama.cpp | 66.58 | 1005.12 | – | – | tg128 | 2026-08-29 | community | – | source |
| Arc Pro B60 | cohere2moe 30B.A3B | IQ4_NL - 4.5 bpw | llama.cpp | 31.31 | 616.59 | – | – | tg128 | 2026-08-29 | community | – | source |
| Arc Pro B60 | cohere2moe 30B.A3B | IQ4_NL - 4.5 bpw | llama.cpp | 31.28 | 531.97 | – | – | tg128 | 2026-08-29 | community | – | source |
| Arc Pro B60 | cohere2moe 30B.A3B | IQ4_NL - 4.5 bpw | llama.cpp | 30.54 | 520.1 | – | – | tg128 | 2026-08-29 | community | – | source |
| Arc Pro B60 | cohere2moe 30B.A3B | IQ4_NL - 4.5 bpw | llama.cpp | 30.51 | 605.7 | – | – | tg128 | 2026-08-29 | community | – | source |