community pool / model answer

DeepSeek-R1-Distill-Qwen-7B

chat · deepseek-ai/DeepSeek-R1-Distill-Qwen-7B

63.6 tok/s · model median · 26 runs in the model record.

Reference rigQuantizationMedian resultBasisTTFT (ms)Peak VRAMMax contextEngines
RTX 2080 8GB4-bit80 tok/sreported-4.7 GB2,048ollama
RTX 3060 12GB4-bit62.1 tok/smeasured1,2474.7 GB131,072llama.cpp, ollama
RTX 3060 Ti 8GB4-bit67.9 tok/smeasured2,1456.7 GB131,072llama.cpp
RTX 3080 12GB4-bit144 tok/sreported-4.8 GB2,048ollama
RTX3060+RTX2080-pooled-20GB 20GB4-bit106.5 tok/sreported-4.8 GB2,048ollama
RX 570 4GB ×24-bit23.1 tok/sreported175- GB2,048llama.cpp

Context tested

The table above is the tested context for each exact rig and quantization cell. Empty fields remain empty rather than being filled with an estimate.