Search results

[Reference interpreter] Remove need for canonicalizing away shape patterns

Quantization

Status: Open.

#2390 In openxla/stablehlo;

· sdasgup3 opened on Jun 11, 2024

Align StableHLO and TOSA arithmetic

Quantization

Spec

Status: Open.

#1898 In openxla/stablehlo;

· sdasgup3 opened on Jan 2, 2024

Consider merging UniformDequantizeOp and UniformQuantizeOp into ConvertOp

Quantization

Spec

Status: Open.

#1576 In openxla/stablehlo;

· burmako opened on Jun 4, 2023

Explore the representation of finer quantization granularities than per-axis

Quantization

Spec

Status: Open.

#1569 In openxla/stablehlo;

· sdasgup3 opened on Jun 2, 2023

Add support for GPTQ style INT3/4/5 quantization of LLMs like LLaMA

Quantization

Spec

Status: Open.

#1491 In openxla/stablehlo;

· powderluv opened on May 16, 2023

Consider introducing QuantizedTensorType based on integer multiplier/shift

Quantization

Spec

Status: Open.

#1404 In openxla/stablehlo;

· sdasgup3 opened on Apr 14, 2023

Spec the behavior of precision_config in DotGeneralOp

Quantization

Spec

Status: Open.

#755 In openxla/stablehlo;

· burmako opened on Dec 13, 2022

Rethink type inference for quantization and sparsity

Integrations

Quantization

Type inference

Status: Open.

#621 In openxla/stablehlo;

· burmako opened on Nov 25, 2022