community pool / model answer

DeepSeek-R1-Distill-Qwen-32B

chat · deepseek-ai/DeepSeek-R1-Distill-Qwen-32B

22.3 tok/s · model median · 4 runs in the model record.

Reference rigQuantizationMedian resultBasisTTFT (ms)Peak VRAMMax contextEngines
RTX 3090 24GB4-bit22.3 tok/smeasured-0 GB8,192llama.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.