implementation detail · filed under optimization
Block-wise FP8 activation quantization with offload
Activations are quantized block-wise in FP8 and combined with offload or remote-offload.
- source
- 1
- model
- 1
- lab adopt it
- 1
- strongest
- used
How sources treat it
One count per evidence span, weakest treatment to strongest.
used 1
Documented in
Evidence
1 span quoted from the sources, strongest treatment first.
In Kimi K3, most activations use block-wise FP8 quantization [58, 30] combined with offload/remote-offload
usedsoftware implementationin Kimi K3Moonshot AI
Filed alongside
Other methods under optimization :: training precision.
NVFP4BF16NVFP4 pre-trainingFP8 mixed-precision trainingFP4+FP8 mixed precisionFP8-precision reinforcement learningMXFP8BF16 gradient reductionBF16 mixed-precision trainingE2M1FP32 attention-output retentionFP32 gradient reductionFP8 storage for the residual stateHigh-precision final network layersMixed-FP8 quantizationMixed-precision trainingNVFP4 fine-grained micro-block scalingTwo-dimensional block quantization