community pool / model answer

Whisper Large v3 Turbo

audio · openai/whisper-large-v3-turbo

39.2 ×real · best measured cell (rtx-3090-24gb) · 3 runs in the model record.

Answer

39.2×real

measuredRTX 3090 24GBfp16 · 3 runs · realtimebest cell in the pool

  1. RTX 3090 24GB39.2 ×realmeasured3 runs · peak VRAM no data yet · 24 GB rig memory
Reference rigQuantizationMedian resultBasisTTFT (ms)Peak VRAMMax contextEngines
RTX 3090 24GBfp1639.2 ×realmeasured-- GB-whisper-large-v3-turbo

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.