Live briefing
Everything about TensorFloat-32
Numbering format in Nvidia hardware
TensorFloat-32 (TF32) is a numeric floating point format designed for Tensor Core running on certain Nvidia GPUs. It was first implemented in the Ampere architecture. TensorFloat-32 combines the 8-bit exponent size of IEEE single precision with the 10-bit mantissa size of half precision for a total of 19 bits per number. It is comparable to the bfloat16 format, which uses a 7-bit mantissa. Wikipedia →
🧠Start here: a 6-step reading path
- Read the overview: TensorFloat-32A 5-minute orientation
- Read the landmark paper: Calculation of cross-correlation function accelerated by TensorFloat-32 Tensor Core operations on NVIDIA’s Ampere and Hopper GPUsCited 8 times
- Go sideways: MinifloatA closely related topic
- Go sideways: Bfloat16 floating-point formatA closely related topic
đź“– Definitions
No dictionary entry for this exact phrase.
🗂️ The subject at a glance
🏆 Landmark papers (most cited)
đź“š Books worth reading
Open Library didn't respond; try again in a moment.
đź”— Related topics
- Minifloat In computing, minifloats are floating-point values represented with very few bits. This reduced precision makes them ill-suited for general-purpose numerical…
- Bfloat16 floating-point format The bfloat16 (brain floating point) floating-point format is a computer number format occupying 16 bits in computer memory; it represents a wide dynamic…
- Mixed-precision arithmetic Mixed-precision arithmetic is a form of floating-point arithmetic that uses numbers with varying widths in a single operation. A common usage of mixed-precision…
- Half-precision floating-point format Half precision (sometimes called FP16 or float16) is a binary floating-point computer number format that occupies 16 bits (two bytes in modern computers)…
- Block floating point Block floating point (BFP) is a method used to provide an arithmetic approaching floating point while using a fixed-point processor. BFP assigns a group…
- Single-precision floating-point format Single-precision floating-point format (sometimes called FP32, float32, or float) is a computer number format, usually occupying 32 bits in computer memory;…
- IEEE 754-1985 IEEE 754-1985 is a historic industry standard for representing floating-point numbers in computers, officially adopted in 1985 and superseded in 2008 by…
- Double-precision floating-point format Double-precision floating-point format (sometimes called FP64 or float64) is a floating-point number format, usually occupying 64 bits in computer memory;…