Data types and precision support

2025-08-28

12 min read time

Applies to Linux and Windows

This topic summarizes the data types supported on AMD GPUs and accelerators and ROCm libraries, along with corresponding HIP data types.

Integral types

The signed and unsigned integral types supported by ROCm are listed in the following table.

Type name	HIP type	Description
int8	int8_t, uint8_t	A signed or unsigned 8-bit integer
int16	int16_t, uint16_t	A signed or unsigned 16-bit integer
int32	int32_t, uint32_t	A signed or unsigned 32-bit integer
int64	int64_t, uint64_t	A signed or unsigned 64-bit integer

Floating-point types

The floating-point types supported by ROCm are listed in the following table.

Type name	HIP type	Description
float4 (E2M1)	__hip_fp4_e2m1	A 4-bit floating-point number with E2M1 bit layout, as described in low precision floating point types page.
float6 (E3M2)	__hip_fp6_e3m2	A 6-bit floating-point number with E3M2 bit layout, as described in low precision floating point types page.
float6 (E2M3)	__hip_fp6_e2m3	A 6-bit floating-point number with E2M3 bit layout, as described in low precision floating point types page.
float8 (E4M3)	__hip_fp8_e4m3_fnuz, __hip_fp8_e4m3	An 8-bit floating-point number with E4M3 bit layout, as described in low precision floating point types page. The FNUZ variant has expanded range with no infinity or signed zero (NaN represented as negative zero), while the OCP variant follows the Open Compute Project specification.
float8 (E5M2)	__hip_fp8_e5m2_fnuz, __hip_fp8_e5m2	An 8-bit floating-point number with E5M2 bit layout, as described in low precision floating point types page. The FNUZ variant has expanded range with no infinity or signed zero (NaN represented as negative zero), while the OCP variant follows the Open Compute Project specification.
float16	half	A 16-bit floating-point number that conforms to the IEEE 754-2008 half-precision storage format.
bfloat16	bfloat16	A shortened 16-bit version of the IEEE 754 single-precision storage format.
tensorfloat32	Not available	A floating-point number that occupies 32 bits or less of storage, providing improved range compared to half (16-bit) format, at (potentially) greater throughput than single-precision (32-bit) formats.
float32	float	A 32-bit floating-point number that conforms to the IEEE 754 single-precision storage format.
float64	double	A 64-bit floating-point number that conforms to the IEEE 754 double-precision storage format.

Note

• The float8 and tensorfloat32 types are internal types used in calculations in Matrix Cores and can be stored in any type of the same size.

• CDNA3 natively supports FP8 FNUZ (E4M3 and E5M2), which differs from the customized FP8 format used with NVIDIA H100 (FP8 Formats for Deep Learning).

• In some AMD documents and articles, float8 (E5M2) is referred to as bfloat8.

• The low precision floating point types page describes how to use these types in HIP with examples.

Level of support definitions

In the following sections, icons represent the level of support. These icons, described in the following table, are also used in the library data type support pages.

Icon	Definition
NA	Not applicable
❌	Not supported
⚠️	Partial support
✅	Full support

Note

• Full support means that the type is supported natively or with hardware emulation.

• Native support means that the operations for that type are implemented in hardware. Types that are not natively supported are emulated with the available hardware. The performance of non-natively supported types can differ from the full instruction throughput rate. For example, 16-bit integer operations can be performed on the 32-bit integer ALUs at full rate; however, 64-bit integer operations might need several instructions on the 32-bit integer ALUs.

• Any type can be emulated by software, but this page does not cover such cases.

Data type support by hardware architecture

AMD’s GPU lineup spans multiple architecture generations:

• CDNA1 such as MI100

• CDNA2 such as MI210, MI250, and MI250X

• CDNA3 such as MI300A, MI300X, and MI325X

• CDNA4 such as MI350X and MI355X

• RDNA2 such as PRO W6800 and PRO V620

• RDNA3 such as RX 7900XT and RX 7900XTX

• RDNA4 such as RX 9070 and RX 9070XT

HIP C++ type implementation support

The HIP C++ types available on different hardware platforms are listed in the following table.

HIP C++ Type	CDNA1	CDNA2	CDNA3	CDNA4	RDNA2	RDNA3	RDNA4
int8_t, uint8_t	✅	✅	✅	✅	✅	✅	✅
int16_t, uint16_t	✅	✅	✅	✅	✅	✅	✅
int32_t, uint32_t	✅	✅	✅	✅	✅	✅	✅
int64_t, uint64_t	✅	✅	✅	✅	✅	✅	✅
__hip_fp4_e2m1	❌	❌	❌	✅	❌	❌	❌
__hip_fp6_e2m3	❌	❌	❌	✅	❌	❌	❌
__hip_fp6_e3m2	❌	❌	❌	✅	❌	❌	❌
__hip_fp8_e4m3_fnuz	❌	❌	✅	❌	❌	❌	❌
__hip_fp8_e5m2_fnuz	❌	❌	✅	❌	❌	❌	❌
__hip_fp8_e4m3	❌	❌	❌	✅	❌	❌	✅
__hip_fp8_e5m2	❌	❌	❌	✅	❌	❌	✅
half	✅	✅	✅	✅	✅	✅	✅
bfloat16	✅	✅	✅	✅	✅	✅	✅
float	✅	✅	✅	✅	✅	✅	✅
double	✅	✅	✅	✅	✅	✅	✅

Note

Library support for specific data types is contingent upon hardware support. Even if a ROCm library indicates support for a particular data type, that type will only be fully functional if the underlying hardware architecture (as shown in the table above) also supports it. For example, fp8 types are only available on architectures shown with a checkmark in the relevant rows.

Compute units support

The following table lists data type support for compute units.

Integral types

Type name	int8	int16	int32	int64
CDNA1	✅	✅	✅	✅
CDNA2	✅	✅	✅	✅
CDNA3	✅	✅	✅	✅
CDNA4	✅	✅	✅	✅
RDNA2	✅	✅	✅	✅
RDNA3	✅	✅	✅	✅
RDNA4	✅	✅	✅	✅

Low precision floating-point types

Type name	float4	float6 (E2M3)	float6 (E3M2)	float8 (E4M3)	float8 (E5M2)
CDNA1	❌	❌	❌	❌	❌
CDNA2	❌	❌	❌	❌	❌
CDNA3	❌	❌	❌	❌	❌
CDNA4	❌	❌	❌	❌	❌
RDNA2	❌	❌	❌	❌	❌
RDNA3	❌	❌	❌	❌	❌
RDNA4	❌	❌	❌	❌	❌

High precision floating-point types

Type name	float16	bfloat16	tensorfloat32	float32	float64
CDNA1	✅	✅	❌	✅	✅
CDNA2	✅	✅	❌	✅	✅
CDNA3	✅	✅	❌	✅	✅
CDNA4	✅	✅	❌	✅	✅
RDNA2	✅	✅	❌	✅	✅
RDNA3	✅	✅	❌	✅	✅
RDNA4	✅	✅	❌	✅	✅

Matrix core support

The following table lists data type support for AMD GPU matrix cores.

Integral types

Type name	int8	int16	int32	int64
CDNA1	✅	❌	❌	❌
CDNA2	✅	❌	❌	❌
CDNA3	✅	❌	❌	❌
CDNA4	✅	❌	❌	❌
RDNA2	✅	❌	❌	❌
RDNA3	✅	❌	❌	❌
RDNA4	✅	❌	❌	❌

Low precision floating-point types

Type name	float4	float6 (E2M3)	float6 (E3M2)	float8 (E4M3)	float8 (E5M2)
CDNA1	❌	❌	❌	❌	❌
CDNA2	❌	❌	❌	❌	❌
CDNA3	❌	❌	❌	✅	✅
CDNA4	✅	✅	✅	✅	✅
RDNA2	❌	❌	❌	❌	❌
RDNA3	❌	❌	❌	❌	❌
RDNA4	❌	❌	❌	✅	✅

High precision floating-point types

Type name	float16	bfloat16	tensorfloat32	float32	float64
CDNA1	✅	✅	❌	✅	❌
CDNA2	✅	✅	❌	✅	✅
CDNA3	✅	✅	✅	✅	✅
CDNA4	✅	✅	✅	✅	✅
RDNA2	✅	✅	❌	❌	❌
RDNA3	✅	✅	❌	❌	❌
RDNA4	✅	✅	❌	❌	❌

Atomic operations support

The following table lists which data types are supported for atomic operations on AMD GPUs. The atomics operation type behavior is affected by the memory locations, memory granularity, or scope of operations. For detailed various support of atomic read-modify-write (atomicRMW) operations collected on the Hardware atomics operation support page.

Integral types

Type name	int8	int16	int32	int64
CDNA1	❌	❌	✅	✅
CDNA2	❌	❌	✅	✅
CDNA3	❌	❌	✅	✅
RDNA3	❌	❌	✅	✅
RDNA4	❌	❌	✅	✅

Low precision floating-point types

Type name	float4	float6 (E2M3)	float6 (E3M2)	float8 (E4M3)	float8 (E5M2)
CDNA1	❌	❌	❌	❌	❌
CDNA2	❌	❌	❌	❌	❌
CDNA3	❌	❌	❌	❌	❌
CDNA4	❌	❌	❌	❌	❌
RDNA2	❌	❌	❌	❌	❌
RDNA3	❌	❌	❌	❌	❌
RDNA4	❌	❌	❌	❌	❌

High precision floating-point types

Type name	2 x float16	2 x bfloat16	tensorfloat32	float32	float64
CDNA1	✅	✅	❌	✅	❌
CDNA2	✅	✅	❌	✅	✅
CDNA3	✅	✅	❌	✅	✅
CDNA4	✅	✅	❌	✅	✅
RDNA2	❌	❌	❌	✅	❌
RDNA3	❌	❌	❌	✅	❌
RDNA4	✅	✅	❌	✅	❌

Note

You can emulate atomic operations using software for cases that are not natively supported. Software-emulated atomic operations have a high negative performance impact when they frequently access the same memory address.

Data type support in ROCm libraries

ROCm library support for int8, float8 (E4M3), float8 (E5M2), int16, float16, bfloat16, int32, tensorfloat32, float32, int64, and float64 is listed in the following tables.

Libraries input/output type support

The following tables list ROCm library support for specific input and output data types. Select a library from the below table to view the supported data types.

Category

ML & Computer Vision

Communication

Math Libraries

Primitives

Library

Composable Kernel

MIGraphX

MIOpen

RCCL

hipBLAS

hipBLASLt

hipFFT

hipRAND

hipSOLVER

hipSPARSE

hipSPARSELt

rocBLAS

rocFFT

rocRAND

rocSOLVER

rocSPARSE

rocWMMA

Tensile

hipCUB

hipTensor

rocPRIM

rocThrust

For more information, please visit Composable Kernel.

Data Type	Support
int8	✅
int32	✅
float4	✅
float6 (E2M3)	✅
float6 (E3M2)	✅
float8 (E4M3)	✅
float8 (E5M2)	✅
float16	✅
bfloat16	✅
float32	✅
float64	✅

For more information, please visit MIGraphX.

Data Type	Support
int8	⚠️
int16	✅
int32	✅
int64	✅
float8 (E4M3)	✅
float8 (E5M2)	✅
float16	✅
bfloat16	✅
float32	✅
float64	✅

For more information, please visit MIOpen.

Data Type	Support
int8	⚠️
int32	⚠️
float8 (E4M3)	⚠️
float8 (E5M2)	⚠️
float16	✅
bfloat16	⚠️
float32	✅
float64	⚠️

For more information, please visit RCCL.

Data Type	Support
int8	✅
int32	✅
int64	✅
float8 (E4M3)	✅
float8 (E5M2)	✅
float16	✅
bfloat16	✅
float32	✅
float64	✅

For more information, please visit hipBLAS.

Data Type	Support
float16	⚠️
bfloat16	⚠️
float32	✅
float64	✅

For more information, please visit hipBLASLt.

Data Type	Support
int8	✅
float4	✅
float6 (E2M3)	✅
float6 (E3M2)	✅
float8 (E4M3)	✅
float8 (E5M2)	✅
float16	✅
bfloat16	✅
float32	✅

For more information, please visit hipFFT.

Data Type	Support
float32	✅
float64	✅

For more information, please visit hipRAND.

Data Type	Support
int8	Output only
int16	Output only
int32	Output only
int64	Output only
float16	Output only
float32	Output only
float64	Output only

For more information, please visit hipSOLVER.

Data Type	Support
float32	✅
float64	✅

For more information, please visit hipSPARSE.

Data Type	Support
float32	✅
float64	✅

For more information, please visit hipSPARSELt.

Data Type	Support
int8	✅
float8 (E4M3)	✅
float8 (E5M2)	✅
float16	✅
bfloat16	✅
float32	✅

For more information, please visit rocBLAS.

Data Type	Support
float16	⚠️
bfloat16	⚠️
float32	✅
float64	✅

For more information, please visit rocFFT.

Data Type	Support
float16	✅
float32	✅
float64	✅

For more information, please visit rocRAND.

Data Type	Support
int8	Output only
int16	Output only
int32	Output only
int64	Output only
float16	Output only
float32	Output only
float64	Output only

For more information, please visit rocSOLVER.

Data Type	Support
float32	✅
float64	✅

For more information, please visit rocSPARSE.

Data Type	Support
float32	✅
float64	✅

For more information, please visit rocWMMA.

Data Type	Support
int8	✅
int32	Output only
float8 (E4M3)	Input only
float8 (E5M2)	Input only
float16	✅
bfloat16	✅
tensorfloat32	✅
float32	✅
float64	✅

For more information, please visit Tensile.

Data Type	Support
int8	✅
int32	✅
float8 (E4M3)	✅
float8 (E5M2)	✅
float16	✅
bfloat16	✅
tensorfloat32	✅
float32	✅
float64	✅

For more information, please visit hipCUB.

Data Type	Support
int8	✅
int16	✅
int32	✅
int64	✅
float16	✅
bfloat16	✅
float32	✅
float64	✅

For more information, please visit hipTensor.

Data Type	Support
float16	✅
bfloat16	✅
float32	✅
float64	✅

For more information, please visit rocPRIM.

Data Type	Support
int8	✅
int16	✅
int32	✅
int64	✅
float16	✅
bfloat16	✅
float32	✅
float64	✅

For more information, please visit rocThrust.

Data Type	Support
int8	✅
int16	✅
int32	✅
int64	✅
float16	⚠️
bfloat16	⚠️
float32	✅
float64	✅

Note

The meaning of partial support depends on the library. Please refer to the individual libraries’ documentation for more information.

Note

As random number generation libraries, rocRAND and hipRAND only specify output data types for the random values they generate, with no need for input data types.

Note

hipBLASLt supports additional data types as internal compute types, which may differ from the supported input/output types shown in the tables above. While TensorFloat32 is not supported as an input or output type in this library, it is available as an internal compute type. For complete details on supported compute types, refer to the hipBLASLt documentation.

hipDataType enumeration

The hipDataType enumeration defines data precision types and is primarily used when the data reference itself does not include type information, such as in void* pointers. This enumeration is mainly utilized in BLAS libraries. The HIP type equivalents of the hipDataType enumeration are listed in the following table with descriptions and values.

hipDataType	HIP type	Value	Description
HIP_R_8I	int8_t	3	8-bit real signed integer.
HIP_R_8U	uint8_t	8	8-bit real unsigned integer.
HIP_R_16I	int16_t	20	16-bit real signed integer.
HIP_R_16U	uint16_t	22	16-bit real unsigned integer.
HIP_R_32I	int32_t	10	32-bit real signed integer.
HIP_R_32U	uint32_t	12	32-bit real unsigned integer.
HIP_R_32F	float	0	32-bit real single precision floating-point.
HIP_R_64F	double	1	64-bit real double precision floating-point.
HIP_R_16F	half	2	16-bit real half precision floating-point.
HIP_R_16BF	bfloat16	14	16-bit real bfloat16 precision floating-point.
HIP_R_8F_E4M3	__hip_fp8_e4m3	28	8-bit real float8 precision floating-point (OCP version).
HIP_R_8F_E5M2	__hip_fp8_e5m2	29	8-bit real bfloat8 precision floating-point (OCP version).
HIP_R_6F_E2M3	__hip_fp6_e2m3	31	6-bit real float6 precision floating-point.
HIP_R_6F_E3M2	__hip_fp6_e3m2	32	6-bit real bfloat6 precision floating-point.
HIP_R_4F_E2M1	__hip_fp4_e2m1	33	4-bit real float4 precision floating-point.
HIP_R_8F_E4M3_FNUZ	__hip_fp8_e4m3_fnuz	1000	8-bit real float8 precision floating-point (FNUZ version).
HIP_R_8F_E5M2_FNUZ	__hip_fp8_e5m2_fnuz	1001	8-bit real bfloat8 precision floating-point (FNUZ version).

The full list of the hipDataType enumeration listed in library_types.h.