community pool / model answer

SmolLM2-135M-Instruct-GGUF

chat · unsloth/SmolLM2-135M-Instruct-GGUF

1,052.6 tok/s · model median · 2 runs in the model record.

Reference rigQuantizationMedian resultBasisTTFT (ms)Peak VRAMMax contextEngines
RTX 5090 32GB2-bit1,102.1 tok/sreported7- GB2,048llama.cpp
RTX 5090 32GB4-bit1,003.1 tok/sreported8- 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.