FrogNano-4B-2609
23 records across 4 hardware.
| hardware | model | quant | backend | tok/s | pp tok/s | W | ttft s | ctx/tg | date | class | flags | source |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| RTX 4090 Laptop GPU | FrogNano-4B-2609 | 4-bit | TensorFold (CUDA) | 319.1 | – | – | – | – | 2026-10-04 | community | – | source |
| RTX 4090 Laptop GPU | FrogNano-4B-2609 | 4-bit | TensorFold (CUDA) | 176.5 | – | – | – | – | 2026-10-04 | community | – | source |
| RTX 4090 Laptop GPU | FrogNano-4B-2609 | 4-bit | TensorFold (CUDA) | 93.4 | – | – | – | – | 2026-10-04 | community | – | source |
| M1 Max | FrogNano-4B-2609 | oQ8e | oMLX | 64.3 | 736.1 | – | – | 4096 | 2026-10-08 | community | – | source |
| M1 Max | FrogNano-4B-2609 | oQ8e | oMLX | 57.8 | 725.2 | – | – | 1024 | 2026-10-08 | community | – | source |
| RTX 4090 Laptop GPU | FrogNano-4B-2609 | 4-bit | TensorFold (CUDA) | 52.4 | – | – | – | – | 2026-10-04 | community | – | source |
| M1 Max | FrogNano-4B-2609 | 4-bit | oMLX | 50.5 | 594.9 | – | – | 8192 | 2026-10-08 | community | – | source |
| RTX 4090 Laptop GPU | FrogNano-4B-2609 | 4-bit | TensorFold (CUDA) | 50.4 | – | – | – | – | 2026-10-04 | community | – | source |
| M4 Pro | FrogNano-4B-2609 | oQ8e | oMLX | 48.6 | 794.8 | – | – | 1024 | 2026-10-10 | community | – | source |
| M4 Pro | FrogNano-4B-2609 | oQ8e | oMLX | 45.0 | 804.7 | – | – | 4096 | 2026-10-10 | community | – | source |
| M4 | FrogNano-4B-2609 | oQ6e | oMLX | 43.4 | 394.5 | – | – | 1024 | 2026-10-09 | community | – | source |
| M4 | FrogNano-4B-2609 | oQ6e | oMLX | 40.4 | 540.1 | – | – | 4096 | 2026-10-09 | community | – | source |
| M4 | FrogNano-4B-2609 | oQ4e | oMLX | 36.2 | 415.3 | – | – | 1024 | 2026-10-07 | community | – | source |
| M4 | FrogNano-4B-2609 | oQ4e | oMLX | 34.6 | 433.1 | – | – | 4096 | 2026-10-07 | community | – | source |
| M1 Max | FrogNano-4B-2609 | oQ8e | oMLX | 31.9 | 273.3 | – | – | 131072 | 2026-10-08 | community | – | source |
| M4 | FrogNano-4B-2609 | gptq | oMLX | 28.5 | 409.4 | – | – | 1024 | 2026-10-08 | community | – | source |
| M4 | FrogNano-4B-2609 | gptq | oMLX | 27.4 | 427.3 | – | – | 4096 | 2026-10-08 | community | – | source |
| M4 | FrogNano-4B-2609 | gptq | oMLX | 27.1 | 394.8 | – | – | 1024 | 2026-10-08 | community | – | source |
| M4 | FrogNano-4B-2609 | gptq | oMLX | 27.0 | 526.5 | – | – | 4096 | 2026-10-08 | community | – | source |
| M4 | FrogNano-4B-2609 | gptq | oMLX | 26.4 | 519.0 | – | – | 4096 | 2026-10-08 | community | – | source |
| M4 | FrogNano-4B-2609 | gptq | oMLX | 26.2 | 393.0 | – | – | 1024 | 2026-10-08 | community | – | source |
| M4 | FrogNano-4B-2609 | oQ8e | oMLX | 21.4 | 412.4 | – | – | 1024 | 2026-10-07 | community | – | source |
| M4 | FrogNano-4B-2609 | oQ8e | oMLX | 21.3 | 423.8 | – | – | 4096 | 2026-10-07 | community | – | source |