G06F 5/012
In floating-point computations
Official title: {in floating-point computations}
G06F5/012 (In floating-point computations) appears on 275 US grants and 255 published applications in this corpus.
Search patents in G06F- Granted patents
- 275
- Published applications
- 255
- Assignees
- 115
- Inventors
- 543
- Latest grant
- 22 Sep 2026
Filings by year
| Year | Granted patents | Published applications |
|---|---|---|
| 1976 | 1 | 0 |
| 1977 | 0 | 0 |
| 1978 | 0 | 0 |
| 1979 | 1 | 0 |
| 1980 | 1 | 0 |
| 1981 | 0 | 0 |
| 1982 | 1 | 0 |
| 1983 | 0 | 0 |
| 1984 | 3 | 0 |
| 1985 | 2 | 0 |
| 1986 | 4 | 0 |
| 1987 | 1 | 0 |
| 1988 | 6 | 0 |
| 1989 | 3 | 0 |
| 1990 | 7 | 0 |
| 1991 | 6 | 0 |
| 1992 | 5 | 0 |
| 1993 | 4 | 0 |
| 1994 | 1 | 0 |
| 1995 | 5 | 0 |
| 1996 | 5 | 0 |
| 1997 | 6 | 0 |
| 1998 | 11 | 0 |
| 1999 | 10 | 0 |
| 2000 | 5 | 0 |
| 2001 | 3 | 0 |
| 2002 | 2 | 11 |
| 2003 | 2 | 3 |
| 2004 | 5 | 5 |
| 2005 | 6 | 3 |
| 2006 | 4 | 6 |
| 2007 | 2 | 1 |
| 2008 | 6 | 3 |
| 2009 | 2 | 0 |
| 2010 | 2 | 1 |
| 2011 | 2 | 1 |
| 2012 | 1 | 4 |
| 2013 | 1 | 3 |
| 2014 | 4 | 8 |
| 2015 | 6 | 12 |
| 2016 | 11 | 18 |
| 2017 | 16 | 13 |
| 2018 | 11 | 17 |
| 2019 | 15 | 13 |
| 2020 | 18 | 20 |
| 2021 | 18 | 10 |
| 2022 | 12 | 24 |
| 2023 | 10 | 19 |
| 2024 | 10 | 15 |
| 2025 | 11 | 24 |
| 2026 | 7 | 21 |
By publication year. Applications have been published since 2001. 2026 runs to 8 Oct.
Recent patents
- Floating point pre-alignment structure for computing-in-memory applications and computing method thereof22 Sep 2026US 12,743,249 B2National Tsing Hua University
- Iterative reduction of quantization error in neural network model1 Sep 2026US 12,725,022 B2Samsung Electronics Co., Ltd.
- High accuracy texture filtering in computer graphics11 Aug 2026US 12,705,817 B2Imagination Technologies Limited
- Multiplication and accumulation(MAC) operator and processing-in-memory (PIM) device including the mac operator26 May 2026US 12,639,041 B2SK hynix Inc.
- Method of performing hardware efficient unbiased rounding of a number5 May 2026US 12,619,394 B2Imagination Technologies Limited
- Multiplication and accumulation (mac) operator10 Sep 2026US 2026/0267610 A1SK hynix Inc.
- Exponential and logarithmic function approximation for floating point numbers27 Aug 2026US 2026/0252664 A1NVIDIA Corporation
- Systems and method for encoding and decoding parameters of a machine learning model13 Aug 2026US 2026/0237101 A1Guangdong Oppo Mobile Telecommunications Corp., Ltd.
- Randomized block-level scaling of neural network tensors30 Jul 2026US 2026/0219841 A1NVIDIA Corporation
- Randomized quantization of neural networks30 Jul 2026US 2026/0220454 A1NVIDIA Corporation
Most cited
By later US patents
- Microprogrammable floating point arithmetic unit capable of performing arithmetic operations on long and short operands211US 4,161,784 AHoneywell Information Systems Inc. · 1979
- Data processor having capability to perform both floating point operations and memory access in response to a single instruction194US 5,673,407 ATexas Instruments Incorporated · 1997
- Numeric data processor119US 4,338,675 AIntel Corporation · 1982
- Fraction bus for use in a numeric data processor100US 4,484,259 AIntel Corporation · 1984
- Normalization method for floating point numbers94US 5,487,022 ATexas Instruments Incorporated · 1996