community pool / model answer

ThinkingCap-Qwen3.6-27B

chat · bottlecapai/ThinkingCap-Qwen3.6-27B

56.1 tok/s · model median · 9 runs in the model record.

Reference rigQuantizationMedian resultBasisTTFT (ms)Peak VRAMMax contextEngines
Arc Pro B70 32GB4-bit27.5 tok/sreported-- GB204,800llama.cpp
Arc Pro B70 32GB5-bit26.3 tok/sreported-- GB131,072llama.cpp
Arc Pro B70 32GB6-bit27.9 tok/sreported-- GB131,072llama.cpp
RTX 3090 24GB ×28-bit58.6 tok/smeasured386- GB131,072llama.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.