List of AMD Ryzen processors

Microprocessors of the Ryzen brand are a line of x86-64 microprocessors manufactured by AMD, based on the Zen microarchitecture. Ryzen microprocessors include: Ryzen 3, Ryzen 5, Ryzen 7, Ryzen 9, and Ryzen Threadripper (with up to 96 cores). All consumer desktop Ryzens (except PRO models), as well as all mobile processors with the HX suffix, have an unlocked multiplier. In addition, all Ryzen microprocessors support Simultaneous Multithreading (SMT), except for earlier Zen/Zen+ based desktop and mobile Ryzen 3 processors, and some models of the Zen 2 mobile Ryzen microprocessor.

Desktop processors

Ryzen 1000 series

Summit Ridge (1000 series, Zen based)

Common features of Ryzen 1000 desktop CPUs:

Socket: AM4.
All the CPUs support DDR4-2666 in dual-channel mode.
All the CPUs support 24 PCIe 3.0 lanes. 4 of the lanes are reserved as link to the chipset.
No integrated graphics.
L1 cache: 96KB (32KB data + 64KB instruction) per core.
L2 cache: 512KB per core.
Node/fabrication process: GlobalFoundries 14 LP.

Branding and Model	Cores (threads)	Clock rate (GHz)	L3 cache (total)	TDP	Core config	Thermal Solution	Release date	Launch price
Base	PBO 1–2 (≥3)	XFR 1–2
Ryzen 7		8 (16)	3.6	4.0 (3.7)	4.1	16MB	95W	2 × 4	Wraith Max	March 2, 2017	US $499
	3.4	3.8 (3.5)	3.9	—N/a	June 29, 2017	OEM
	Wraith Max	March 2, 2017	US $399
	3.0	3.7 (3.2)	3.75	65W	—N/a	June 29, 2017	OEM
	Wraith Spire	March 2, 2017	US $329
Ryzen 5		6 (12)	3.6	4.0 (3.7)	4.1	95W	2 × 3	Wraith Max	April 11, 2017	US $249
	3.2	3.6 (3.4)	3.7	65W	—N/a	June 29, 2017	OEM
	Wraith Stealth	April 11, 2017	US $219
	4 (8)	3.5	3.7 (3.6)	3.9	2 × 2	Wraith Spire (No LED)	US $189
	—N/a	June 29, 2017	OEM
	3.2	3.4 (3.4)	3.45	8MB	Wraith Stealth	April 11, 2017	US $169
Ryzen 3		4 (4)	3.5	3.7 (3.5)	3.9	July 27, 2017	US $129
	—N/a	June 29, 2017	OEM
	3.1	3.4 (3.1)	3.45
	Wraith Stealth	July 27, 2017	US $109

Whitehaven (Threadripper 1000 series, Zen based)

Common features of Ryzen 1000 HEDT CPUs:

Socket: TR4.
All the CPUs support DDR4-2666 in quad-channel mode.
All the CPUs support 64 PCIe 3.0 lanes. 4 of the lanes are reserved as link to the chipset.
No integrated graphics.
L1 cache: 96KB (32KB data + 64KB instruction) per core.
L2 cache: 512KB per core.
Node/fabrication process: GlobalFoundries 14LP.

Branding and Model	Cores (threads)	Clock rate (GHz)	L3cache (total)	TDP	Chiplets	Core config	Release date	Launch price
Base	PBO 1–4 (≥5)	XFR 1–2
Ryzen Threadripper		16 (32)	3.4	4.0 (3.7)	4.2	32MB	180W	2 × CCD	4 × 4	August 10, 2017	US $999
	12 (24)	3.5	4 × 3	US $799
	8 (16)	3.8	4.0 (3.9)	16MB	2 × 4	August 31, 2017	US $549

Ryzen 2000 series

Raven Ridge (2000 series with Radeon Graphics, Zen/GCN5 based)

Common features of Ryzen 2000 desktop APUs:

Socket: AM4.
All the CPUs support DDR4-2933 in dual-channel mode.
All the CPUs support 16 PCIe 3.0 lanes. 4 of the lanes are reserved as link to the chipset.
Includes integrated GCN 5th generation GPU.
L1 cache: 96KB (32KB data + 64KB instruction) per core.
L2 cache: 512KB per core.
Fabrication process: GlobalFoundries 14LP.

Branding and Model	CPU	GPU	TDP	Release date	Launch MSRP
Cores (threads)	Clock rate (GHz)	L3 cache (total)	Model	Clock (GHz)	Config	Processing power (GFLOPS)
Base	Boost
Ryzen 5		4 (8)	3.6	3.9	4 MB	RX Vega 11	1.25	704:44:16 11 CU	1760	46–65W	February 12, 2018	US $169
	3.2	3.8	35W	April 19, 2018	OEM
Ryzen 3		4 (4)	3.5	3.7	Vega 8	1.1	512:32:16 8 CU	1126	46–65W	February 12, 2018	US $99
	3.2	3.6	35W	April 19, 2018	OEM
PRO 2100GE	2 (4)	—N/a	Vega 3	1.0	192:12:4 3 CU	384	2019

Pinnacle Ridge (2000 series, Zen+ based)

Common features of Ryzen 2000 desktop CPUs:

Socket: AM4.
All the CPUs support DDR4-2933 in dual-channel mode, except for R7 2700E, R5 2600E, R5 1600AF and R3 1200AF which support it at DDR4-2666 speeds.
All the CPUs support 24 PCIe 3.0 lanes. 4 of the lanes are reserved as link to the chipset.
No integrated graphics.
L1 cache: 96KB (32KB data + 64KB instruction) per core.
L2 cache: 512KB per core.
Fabrication process: GlobalFoundries 12LP (14LP+).

Branding and Model	Cores (threads)	Clock rate (GHz)	L3 cache (total)	TDP	Core config	Thermal Solution	Release date	Launch price
Base	PB2
Ryzen 7		8 (16)	3.7	4.3	16 MB	105W	2 × 4	Wraith Prism	April 19, 2018	US $329
	3.2	4.1	65 W	Wraith Spire	US $299
	2.8	4.0	45W	—N/a	September 19, 2018	OEM
Ryzen 5		6 (12)	3.6	4.2	95W	2 × 3	Wraith Spire	April 19, 2018	US $229
	3.4	3.9	65W	Wraith Stealth	US $199
	3.1	4.0	45W	—N/a	September 19, 2018	OEM
	3.2	3.6	65W	Wraith Stealth	October 11, 2019	US $85
	4 (8)	3.6	4.0	8 MB	1 × 4	—N/a	September 10, 2018	OEM
Ryzen 3		4 (4)	3.5
1200(AF)	3.1	3.4	Wraith Stealth	April 21, 2020	US $60

Colfax (Threadripper 2000 series, Zen+ based)

Common features of Ryzen 2000 HEDT CPUs:

Socket: TR4.
All the CPUs support DDR4-2933 in quad-channel mode.
All the CPUs support 64 PCIe 3.0 lanes. 4 of the lanes are reserved as link to the chipset.
No integrated graphics.
L1 cache: 96KB (32KB data + 64KB instruction) per core.
L2 cache: 512KB per core.
Fabrication process: GlobalFoundries 12LP (14LP+).

Branding and Model	Cores (threads)	Clock rate (GHz)	L3 cache (total)	TDP	Chiplets	Core config	Release date	Launch price
Base	PB2
Ryzen Threadripper		32 (64)	3.0	4.2	64MB	250W	4 × CCD	8 × 4	Aug 13, 2018	US $1799
	24 (48)	8 × 3	Oct 2018	US $1299
	16 (32)	3.5	4.4	32MB	180W	2 × CCD	4 × 4	Aug 31, 2018	US $899
	12 (24)	4.3	4 × 3	Oct 2018	US $649

Ryzen 3000 series

Picasso (3000 series with Radeon Graphics, Zen+/GCN5 based)

Common features of Ryzen 3000 desktop APUs:

Socket: AM4.
All the CPUs support DDR4-2933 in dual-channel mode.
L1 cache: 96KB (32KB data + 64KB instruction) per core.
L2 cache: 512KB per core.
All the CPUs provide 20 PCIe 3.0, from which 4 are reserved as link to the chipset.
Includes integrated GCN 5th generation GPU.
Fabrication process: GlobalFoundries 12LP (14LP+).

Branding and Model	CPU	GPU	TDP	Release date	MSRP
Cores (threads)	Clock rate (GHz)	L3 cache (total)	Model	Clock (GHz)	Config	Processing power (GFLOPS)
Base	Boost
Ryzen 5		4 (8)	3.7	4.2	4MB	Radeon RXVega11	1.4	704:44:8 11 CU	1971.2	65W	Sep 30, 2019	OEM
	Jul 7, 2019	US$149
	3.3	4.0	1.3	1830.4	35W	Sep 30, 2019	OEM
	3.6	Radeon Vega10	640:40:8 10 CU	1664	65W	Jul 21, 2020
	4 (4)	3.3	3.9	1.2	1536	35W
Ryzen 3		3.6	4.0	Radeon Vega8	1.25	512:32:8 8 CU	1280	65W	Sep 30, 2019
	Jul 7, 2019	US$99
	3.3	3.8	1.2	1228.8	35W	Jul 7, 2019	OEM
	Sep 30, 2019

Matisse (3000 series, Zen 2 based)

Common features of Ryzen 3000 desktop CPUs:

Socket: AM4.
All the CPUs support DDR4-3200 in dual-channel mode.
L1 cache: 64KB (32KB data + 32KB instruction) per core.
L2 cache: 512KB per core.
All the CPUs support 24 PCIe 4.0 lanes. 4 of which are reserved as link to the chipset.
No integrated graphics.
Fabrication process: TSMC 7FF.

Branding and Model	Cores (threads)	Clock rate (GHz)	L3 cache (total)	TDP	Chiplets	Core config	Thermal Solution	Release date	MSRP
Base	Boost
Ryzen 9		16 (32)	3.5	4.7	64 MB	105W	2 × CCD 1 × I/OD	4 × 4	—N/a	Nov 25, 2019	US $749
	12 (24)	3.8	4 × 3	Jul 7, 2020	US $499
	4.6	Wraith Prism	Jul 7, 2019
	3.1	4.3	65 W	—N/a	Oct 8, 2019	OEM
Ryzen 7		8 (16)	3.9	4.7	32 MB	105W	1 × CCD 1 × I/OD	2 × 4	—N/a	Jul 7, 2020	US $399
	4.5	Wraith Prism	Jul 7, 2019
	3.6	4.4	065W	US $329
Ryzen 5		6 (12)	3.8	4.5	95 W	2 × 3	—N/a	Jul 7, 2020	US $249
	4.4	Wraith Spire (non-LED)	Jul 7, 2019
	3.6	4.2	65 W	Wraith Stealth	US $199
	6 (6)	4.1	Oct 8, 2019	CNY 1099 (Mainland China Only)
	16 MB	—N/a	Nov 15, 2019	OEM (Worldwide) JPY 16000 (Japan Only)
Ryzen 3		4 (8)	3.8	4.3	1 × 4	Wraith Stealth	Apr 21, 2020	US $119
	3.6	3.9	2 × 2	US $99

Castle Peak (Threadripper 3000 series, Zen 2 based)

Common features of Ryzen 3000 HEDT/workstation CPUs:

Socket: sTRX4 (Threadripper), sWRX8 (Threadripper PRO).
Threadripper CPUs support DDR4-3200 in quad-channel mode while Threadripper PRO CPUs support DDR4-3200 in octa-channel mode.
L1 cache: 64KB (32KB data + 32KB instruction) per core.
L2 cache: 512KB per core.
Threadripper CPUs support 64 PCIe 4.0 lanes while Threadripper PRO CPUs support 128 PCIe 4.0 lanes. 8 of the lanes are reserved as link to the chipset.
No integrated graphics.
Fabrication process: TSMC 7FF.

Branding and Model	Cores (threads)	Clock rate (GHz)	L3 cache (total)	TDP	Chiplets	Core config	Release date	MSRP
Base	Boost
Ryzen Threadripper PRO		64 (128)	2.7	4.2	256 MB	280W	8 × CCD 1 × I/OD	16 × 4	Jul 14, 2020
	32 (64)	3.5	128 MB	4 × CCD 1 × I/OD	8 × 4
	16 (32)	3.9	4.3	64 MB	2 × CCD 1 × I/OD	4 × 4
	12 (24)	4.0	4 × 3
Ryzen Threadripper		64 (128)	2.9	256 MB	8 × CCD 1 × I/OD	16 × 4	Feb 7, 2020	US $3990
	32 (64)	3.7	4.5	128 MB	4 × CCD 1 × I/OD	8 × 4	Nov 25, 2019	US $1999
	24 (48)	3.8	8 × 3	US $1399

Ryzen 4000 series

Renoir (4000 series, Zen 2 based)

Based on the Ryzen 4000G series APUs with the integrated GPU disabled. Common features of Ryzen 4000 desktop CPUs:

Socket: AM4. The AMD 4700S uses the BGA 2197 socket and the 4800S uses BGA 2963.
All the CPUs support DDR4-3200 in dual-channel mode.
L1 cache: 64KB (32KB data + 32KB instruction) per core.
L2 cache: 512KB per core.
All the CPUs support 24 PCIe 3.0 lanes. 4 of the lanes are reserved as link to the chipset.
No integrated graphics.
Fabrication process: TSMC 7FF.
Bundled with AMD Wraith Stealth

The AMD 4700S and 4800S desktop processors are part of a "desktop kit" that comes bundled with a motherboard and GDDR6 RAM. The CPU is soldered, and provides 4 PCIe 2.0 lanes. These are reportedly cut-down variants of the APUs found on the PlayStation 5 and Xbox Series X and S repurposed from defective chip stock.

Branding and model	Cores (threads)	Clock rate (GHz)	L3 cache (total)	TDP	Core config	Release date	MSRP
Base	Boost
AMD	4800S	8 (16)		4.0	8MB		2 × 4	2022	bundled with desktop kit
	3.6	75W	2021
Ryzen 5		6 (12)	4.1	65W	2 × 3	Apr 4, 2022	US $129
Ryzen 3		4 (8)	3.8	4.0	4MB	1 × 4	US $99

Renoir (4000 series with Radeon Graphics, Zen 2/GCN5.1 based)

Common features of Ryzen 4000 desktop APUs:

Socket: AM4.
All the CPUs support DDR4-3200 in dual-channel mode.
L1 cache: 64KB (32KB data + 32KB instruction) per core.
L2 cache: 512KB per core.
All the CPUs support 24 PCIe 3.0 lanes. 4 of the lanes are reserved as link to the chipset.
Includes integrated GCN 5th generation GPU.
Fabrication process: TSMC 7FF.

Branding and model	CPU	GPU	TDP	Release date	Release price
Cores (threads)	Clock rate (GHz)	L3 cache (total)	Core Config	Model	Clock (GHz)	Config	Processing power (GFLOPS)
Base	Boost
Ryzen 7		8 (16)	3.6	4.4	8MB	2 × 4	Radeon Graphics	2.1	512:32:16 8 CU	2150.4	65W	Jul 21, 2020	OEM
	3.1	4.3	2.0	2048	35W
Ryzen 5		6 (12)	3.7	4.2	2 × 3	1.9	448:28:14 7 CU	1702.4	65W	Jul 21, 2020 (OEM) Apr 4, 2022 (retail)	US$154
	3.3	35W	Jul 21, 2020	OEM
Ryzen 3		4 (8)	3.8	4.0	4MB	1 × 4	1.7	384:24:12 6 CU	1305.6	65W
	3.5	35W

Ryzen 5000 series

Vermeer (5000 series, Zen 3 based)

Common features of Ryzen 5000 desktop CPUs:

Socket: AM4.
All the CPUs support DDR4-3200 in dual-channel mode.
All the CPUs support 24 PCIe 4.0 lanes. 4 of the lanes are reserved as link to the chipset.
No integrated graphics.
L1 cache: 64KB per core (32KB data + 32KB instruction).
L2 cache: 512KB per core.
Fabrication process: TSMC 7FF.

Branding and model	Cores (threads)	Clock rate (GHz)	L3 cache (total)	TDP	Chiplets	Core config	Thermal solution	Release date	MSRP
Base	Boost
Ryzen 9		16 (32)	3.4	4.9	64MB	105W	2 × CCD 1 × I/OD	2 × 8	—N/a	Nov 5, 2020	US$799
	3.3	4.8	Jul 31, 2024	US$349
	12 (24)	3.7	2 × 6	Nov 5, 2020	US$549
	3.0	4.7	65W	Jan 12, 2021	OEM
	Sep 2022
Ryzen 7		8 (16)	3.4	4.5	96MB	105W	1 × CCD 1 × I/OD	1 × 8	Apr 20, 2022 Jun 25, 2026 (10th ann. ed.)	US$449 US$349 (10th ann. ed.)
	3.8	4.8	32MB	Wraith Prism, None	Jul 31, 2024	US$249
	4.7	—N/a	Nov 5, 2020	US$449
	3.4	4.6	65W	Jan 12, 2021	OEM
	3.0	4.1	96MB	105W	Jan 31, 2024	US $249
	3.4	4.6	32 MB	65 W	Apr 4, 2022	US$299
	Sep 2022	OEM
Ryzen 5		6 (12)	3.3	4.4	96MB	105W	1 × 6	Jul 7, 2023	US$229 US Only
	3.7	4.7	32MB	65W	Wraith Stealth	Oct 31, 2024	US$194
	3.7	4.6	Nov 5, 2020	US$299
	3.5	4.5	Oct 31, 2024	US$186
	3.5	4.4	Apr 4, 2022	US$199
	3.0	4.0	Sep 16, 2025	APJ Only
	3.7	4.6	—N/a	Sep 2022	OEM
	3.0	4.0	96MB	105W	Jun 5, 2025	¥‌1099, LATAM, China and UK Only

Cezanne (5000 series, Zen 3 based)

Cézanne based CPUs that have the integrated GPU disabled. Common features of Ryzen 5000 (Cezanne) desktop CPUs:

Socket: AM4.
CPUs support DDR4-3200 in dual-channel mode.
L1 cache: 64KB (32KB data + 32KB instruction) per core.
L2 cache: 512KB per core.
CPUs support 24 PCIe 3.0 lanes. 4 of the lanes are reserved as link to the chipset.
No integrated graphics.
Fabrication process: TSMC 7FF.

Branding and model	Cores (threads)	Thermal solution	Clock rate (GHz)	L3 cache (total)	TDP	Core config	Release date	MSRP (USD)
Base	Boost
Ryzen7		8 (16)	Wraith Stealth	3.7	4.6	16MB	65W	1 × 8	Apr 4, 2022 (OEM), Jan 31, 2024 (retail)	$175
Ryzen5		6 (12)	3.6	4.2	1 × 6	Apr 4, 2022	$159
Ryzen3	5100	4 (8)	–	3.8	8MB	1 × 4	2023	OEM

Cezanne (5000 series with Radeon Graphics, Zen 3/GCN5.1 based)

Common features of Ryzen 5000 desktop APUs:

Socket: AM4.
All the CPUs support DDR4-3200 in dual-channel mode.
L1 cache: 64KB (32KB data + 32KB instruction) per core.
L2 cache: 512KB per core.
All the CPUs support 24 PCIe 3.0 lanes. 4 of the lanes are reserved as link to the chipset.
Includes integrated GCN 5th generation GPU.
Fabrication process: TSMC 7FF.

Branding and model	CPU	GPU	Thermal solution	TDP	Release date	MSRP
Cores (threads)	Clock rate (GHz)	L3 cache (total)	Core config	Clock (MHz)	Config	Processing power (GFLOPS)
Base	Boost
Ryzen 7		8 (16)	3.8	4.6	16MB	1 × 8	2000	512:32:8 8CU	2048	—N/a	65W
	Wraith Stealth	Apr 13, 2021(OEM), Aug 5, 2021(retail)	US$359
	3.2	—N/a	35W
	Wraith Stealth	Apr 13, 2021	OEM
Ryzen 5		6 (12)	3.6	1 × 6	1900	448:28:8 7CU	1702.4	65W	Jan 31, 2024	US$140
	3.9	4.4	—N/a
	Wraith Stealth	Apr 13, 2021(OEM), Aug 5, 2021(retail)	US$259
	3.4	—N/a	35W
	Wraith Stealth	Apr 13, 2021	OEM
	3.6	65W	Jan 31, 2024	US$125
Ryzen 3		4 (8)	4.0	4.2	8MB	1 × 4	1700	384:24:8 6CU	1305.6	—N/a
	OEM	Apr 13, 2021	OEM
	3.6	—N/a	35W
	OEM	Apr 13, 2021	OEM

Chagall (Threadripper 5000 series, Zen 3 based)

Common features of Ryzen 5000 workstation CPUs:

Socket: sWRX8.
All the CPUs support DDR4-3200 in octa-channel mode.
L1 cache: 64KB (32KB data + 32KB instruction) per core.
L2 cache: 512KB per core.
All the CPUs support 128 PCIe 4.0 lanes. 8 of the lanes are reserved as link to the chipset.
No integrated graphics.
Fabrication process: TSMC 7FF.

Branding and Model	Cores (threads)	Clock rate (GHz)	L3 cache (total)	TDP	Chiplets	Core config	Release date	MSRP
Base	Boost
Ryzen Threadripper PRO		64 (128)	2.7	4.5	256MB	280W	8 × CCD 1 × I/OD	8 × 8	Mar 8, 2022 (OEM) / ? (retail)	OEM / US $6500
	32 (64)	3.6	128MB	4 × CCD 1 × I/OD	4 × 8	Mar 8, 2022 (OEM) / ? (retail)	OEM / US $3300
	24 (48)	3.8	4 × 6	Mar 8, 2022 (OEM) / ? (retail)	OEM / US $2400
	16 (32)	4.0	64MB	2 × CCD 1 × I/OD	2 × 8	Mar 8, 2022	OEM
	12 (24)	4.1	2 × 6

Ryzen 7000 series

Raphael (7000 series, Zen 4/RDNA2 based)

Common features of Ryzen 7000 desktop CPUs:

Socket: AM5.
All the CPUs support DDR5-5200 RAM in dual-channel mode in 2x1R and 2x2R configuration, but only DDR5-3600 for 4x1R and 4x2R.
All the CPUs support 28 PCIe 5.0 lanes. 4 of the lanes are reserved as link to the chipset.
Includes integrated RDNA 2 GPU on the I/O die with 2 CUs and clock speeds of 400MHz (base), 2.2GHz (boost). Models with "F" suffixes are without iGPUs.
L1 cache: 64KB (32KB data + 32KB instruction) per core.
L2 cache: 1MB per core.
Fabrication process: TSMC N5 FinFET (N6 FinFET for the I/O die).

Branding and model	Cores (threads)	Clock rate (GHz)	L3 cache (total)	Chiplets	Core config	TDP	Thermal solution	Release date	MSRP
Base	Boost
Ryzen 9		16 (32)	4.2	5.7	128MB	2 × CCD 1 × I/OD	2 × 8	120W	—N/a	Feb 28, 2023	US $699
	4.5	64MB	170W	Sep 27, 2022
	12 (24)	4.4	5.6	128MB	2 × 6	120W	Feb 28, 2023	US $599
	4.7	64MB	170W	Sep 27, 2022	US $549
	3.7	5.4	65W	Wraith Prism, None	Jan 10, 2023	US $429
	Wraith Spire, Wraith Stealth	Jun 13, 2023	OEM
Ryzen 7		8 (16)	4.2	5.0	96MB	1 × CCD 1 × I/OD	1 × 8	120W	—N/a	Apr 6, 2023	US $449
	4.0	4.5	Jul 16, 2026	US $329
	4.5	5.4	32MB	105W	Sep 27, 2022	US $399
	3.8	5.3	65W	Wraith Prism, None	Jan 10, 2023	US $329
	Wraith Spire, Wraith Stealth	Jun 13, 2023	OEM
Ryzen 5		6 (12)	4.1	4.7	96MB	1 × 6	—N/a	Aug 31, 2024	US $299
	4.7	5.3	32MB	105W	Sep 27, 2022
	3.8	5.1	65W	Wraith Stealth	Jan 10, 2023	US $229
	Wraith Spire, Wraith Stealth	Jun 13, 2023	OEM
	4.0	4.5	96MB	—N/a	Nov 12, 2025	OEM
	3.7	5.0	32MB	Wraith Stealth	Jul 22, 2023	US $179
	4.7	Jan 9, 2025	CNY 849 (Mainland China) APJ Only
	3.3	4.3	16MB	Sep 16, 2025	TBA

Storm Peak (Threadripper 7000 series, Zen 4 based)

Common features of Ryzen 7000 HEDT/workstation CPUs:

Socket: sTR5.
Threadripper CPUs support DDR5-5200 in quad-channel mode while Threadripper PRO CPUs support DDR5-5200 in octa-channel mode with ECC support.
L1 cache: 64KB (32KB data + 32KB instruction) per core.
L2 cache: 1MB per core.
Threadripper CPUs support 48 PCIe 5.0 and 24 PCIe 4.0 lanes while Threadripper PRO CPUs support 128 PCIe 5.0 lanes. In addition, all processor models have 4 PCIe 4.0 lanes reserved as link to the chipset.
No integrated graphics.
Fabrication process: TSMC 5FF.

Branding and model	Cores (threads)	Clock rate (GHz)	L3 cache (total)	TDP	Chiplets	Core config	Release date	MSRP
Base	Boost
Ryzen Threadripper PRO		96 (192)	2.5	5.1	384MB	350W	12 × CCD 1 × I/OD	12 × 8	Nov 21, 2023	US $9999[citation needed]
	64 (128)	3.2	256MB	8 × CCD 1 × I/OD	8 × 8	US $7349[citation needed]
	32 (64)	4.0	5.3	128MB	4 × CCD 1 × I/OD	4 × 8	US $3899[citation needed]
	24 (48)	4.2	4 × 6	US $2649[citation needed]
	16 (32)	4.5	64MB	2 × CCD 1 × I/OD	2 × 8	US $1899[citation needed]
	12 (24)	4.7	2 × 6	US $1399[citation needed]
Ryzen Threadripper		64 (128)	3.2	5.1	256MB	8 × CCD 1 × I/OD	8 × 8	US $4999
	32 (64)	4.0	5.3	128MB	4 × CCD 1 × I/OD	4 × 8	US $2499
	24 (48)	4.2	4 × 6	US $1499

Ryzen 8000 series

Phoenix (8000 series, Zen 4 based)

Common features of Ryzen 8000 desktop CPUs:

Socket: AM5.
All the CPUs support DDR5-5200 RAM in dual-channel mode in 2x1R and 2x2R configuration, but only DDR5-3600 for 4x1R and 4x2R.
L1 cache: 64KB (32KB data + 32KB instruction) per core.
L2 cache: 1MB per core.
All the CPUs support 20 PCIe 4.0 lanes. 4 of the lanes are reserved as link to the chipset.
No integrated graphics.
Fabrication process: TSMC 4 nm FinFET.

Branding and model	Cores (threads)	Clock rate (GHz)	L3 cache (total)	NPU	TDP	Core config	Thermal solution	Release date	MSRP
Base	Boost
Ryzen 7		8 (16)	4.1	5.0	16MB	Yes	65W	1 × 8	Wraith Stealth	April 1, 2024	OEM
Ryzen 5		6 (12)	4.2	4.7	No	1 × 6

Phoenix (8000 series with Radeon Graphics, Zen 4/RDNA3/XDNA based)

Common features of Ryzen 8000G desktop APUs:

Socket: AM5.
All the CPUs support DDR5-5200 RAM in dual-channel mode in 2x1R and 2x2R configuration, but only DDR5-3600 for 4x1R and 4x2R.
L1 cache: 64KB (32KB data + 32KB instruction) per core.
L2 cache: 1MB per core.
Models with Zen 4c cores (codenamed Phoenix 2) support 14 PCIe 4.0 lanes, while models without them support 20 lanes. 4 of the lanes are reserved as link to the chipset.
Includes integrated RDNA 3 GPU.
Includes XDNA AI Engine (Ryzen AI) on models without Zen 4c cores.
Fabrication process: TSMC 4 nm FinFET.

Branding and model	CPU	GPU	TDP	Thermal solution	Release date	MSRP
Cores (threads)	Clock rate (GHz)	L3 cache (total)	Core config	Model	Clock (GHz)	Processing power (GFLOPS)	Ryzen AI
Total	Zen 4	Zen 4c	Base	Boost
Ryzen 7		8 (16)	8 (16)	—N/a	4.2	5.1	16MB	1 × 8	780M 12 CU	2.9	4513	Yes	65W	Wraith Spire (before Aug 1, 2025) Wraith Stealth (since Aug 1, 2025)	Jan 31, 2024	US $329
	3.6	2.7	4201	35W	—N/a	Apr 16, 2024[citation needed]	US $299
Ryzen 5		6 (12)	6 (12)	4.3	5.0	1 × 6	760M 8 CU	2.8	2868	65W	Wraith Stealth	Jan 31, 2024	US $229
	3.9	2.6	2663	35 W	—N/a	Apr 16, 2024[citation needed]	OEM
	2 (4)	4 (8)	4.1 / 3.2	5.0 / 3.7	2 + 4	740M 4 CU	2.8	1434	No	65W	Wraith Stealth	Jan 31, 2024	US $179
	3.9 / 3.1	35 W	—N/a	Apr 16, 2024[citation needed]	OEM
Ryzen 3		4 (8)	1 (2)	3 (6)	4.0 / 3.2	4.9 / 3.6	8MB	1 + 3	2.6	1331	65 W	Wraith Stealth	Jan 2024 (OEM) / Q1 2024 (retail)	OEM / TBA
	35 W	—N/a	Apr 16, 2024[citation needed]	OEM

Ryzen 9000 series

Granite Ridge (9000 series, Zen 5/RDNA2 based)

Common features of Ryzen 9000 desktop CPUs:

Socket: AM5.
All the CPUs support DDR5-5600 RAM in dual-channel mode in 2x1R and 2x2R configuration, but only DDR5-3600 for 4x1R and 4x2R.
All the CPUs support 28 PCIe 5.0 lanes. 4 of the lanes are reserved as link to the chipset.
Includes integrated RDNA 2 GPU with 2 CUs and base and boost clock speeds of 0.4 GHz and 2.2 GHz, respectively. Models with "F" suffixes are without iGPUs.
L1 cache: 80 KB (48 KB data + 32 KB instruction) per core.
L2 cache: 1 MB per core.
Models with an "X3D" suffix include 2nd-generation 3D V-Cache in the form of a 64 MB cache die, which augments the L3 cache of one CCD.
Models with an "X3D2" suffix or branded as "Dual Edition" include 2nd-generation 3D V-Cache in the form of two 64 MB cache dies (totaling to 128 MB), one die per CCD, which augment the L3 cache of their respective CCD.
Fabrication process: TSMC N4X FinFET (N6 FinFET for the I/O die).

Branding and Model	Cores (threads)	Clock rate (GHz)	L3 cache (total)	TDP	Chiplets	Core config	Thermal solution	Release date	Launch MSRP
Base	Boost
Ryzen 9		16 (32)	4.3	5.6	192 MB	200 W	2 × CCD 1 × I/OD	2 × 8	—N/a	April 22, 2026	US $899
	5.7	128 MB	170 W	March 12, 2025	US $699
	5.5	June 30, 2026	OEM
	64 MB	June 30, 2026	OEM
	5.7	August 15, 2024	US $649
	12 (24)	3.4	5.4	120 W	2 × 6	June 30, 2026	OEM
	65 W	Wraith Stealth	September 16, 2025	OEM
	4.4	5.5	128 MB	120 W	—N/a	March 12, 2025	US $599
	5.6	64 MB	August 15, 2024	US $499
Ryzen 7		8 (16)	4.7	96 MB	1 × CCD 1 × I/OD	1 × 8	January 29, 2026	US $499
	5.2	November 7, 2024	US $479
	3.8	5.4	32 MB	120 W	June 30, 2026	OEM
	65 W	Wraith Stealth	September 16, 2025	OEM
	5.5	65 W	—N/a	August 8, 2024	US $359
	65 W	September 16, 2025	TBA
Ryzen 5		6 (12)	3.9	5.4	120 W	1 × 6	June 30, 2026	OEM
	65 W	August 8, 2024	US $279
	3.8	5.2	65 W	Wraith Stealth	February 19, 2025	TBA
	5.0	September 16, 2025	CN ¥1,299

Shimada Peak (Threadripper 9000 series, Zen 5 based)

Common features of Ryzen 9000 HEDT/workstation CPUs:

Socket: sTR5.
Threadripper CPUs support DDR5-6400 in quad-channel mode while Threadripper PRO CPUs support DDR5-6400 in octa-channel mode with ECC support.
L1 cache: 80KB (48KB data + 32KB instruction) per core.
L2 cache: 1MB per core.
Threadripper CPUs support 80 PCIe 5.0 lanes (4 of which runs PCIe 4.0 mode to the chipset; and remaining are usually running as 48 PCIe 5.0 lanes plus 24 PCIe 4.0 lanes), while Threadripper PRO CPUs support 128 PCIe 5.0 lanes (4 of which runs in PCIe 4.0 mode to the chipset), and extra 8 PCIe 3.0 lanes. Some lanes support alternative mode of running SATA3.
No integrated graphics.
Fabrication process: TSMC 4nm FinFET.

Branding and model	Cores (threads)	Clock rate (GHz)	L3 cache (total)	TDP	Chiplets	Core config	Release date	MSRP
Base	Boost
Ryzen Threadripper PRO		96 (192)	2.5	5.4	384MB	350W	12 × CCD 1 × I/OD	12 × 8	July 2025	$11,699
	64 (128)	3.2	256MB	8 × CCD 1 × I/OD	8 × 8	$7,999
	32 (64)	4.0	128MB	4 × CCD 1 × I/OD	4 × 8	$4,099
	24 (48)	4.2	4 × 6	$2,899
	16 (32)	4.5	64MB	2 × CCD 1 × I/OD	2 × 8	$1,649
	12 (24)	4.7	2 × 6	OEM
Ryzen Threadripper		64 (128)	3.2	256MB	8 × CCD 1 × I/OD	8 × 8	$4,999
	32 (64)	4.0	128MB	4 × CCD 1 × I/OD	4 × 8	$2,499
	24 (48)	4.2	4 × 6	$1,499

Ryzen AI 400 Series

Gorgon Point AM5 (AI 400 Series, Zen 5/RDNA3.5/XDNA2 based)

Common features of Ryzen AI 400 desktop APUs:

Socket: AM5.
All the CPUs support DDR5-5600 RAM in dual-channel mode in 2x1R and 2x2R configuration, but only DDR5-3600 for 4x1R and 4x2R.
L1 cache: 80 KB per core.
L2 cache: 1 MB per core.
Includes integrated RDNA 3.5 GPU.
Includes XDNA 2 AI Engine (Ryzen AI).
Fabrication process: TSMC 4 nm FinFET.

Branding and model	CPU	GPU	NPU (Ryzen AI)	TDP	Release	MSRP
Cores (threads)	Clock rate (GHz)	L3 cache
Base	Boost	Model	CUs	Clock (GHz)
Ryzen AI 7	()	8 (16)	2.0	5.1	16 MB	Radeon 860M	8	3.1	Up to 50 TOPS	65 W	March 2, 2026 (OEM)	OEM
()	35 W
Ryzen AI 5	()	6 (12)	4.8	16 MB	Radeon 840M	4	2.9	65 W
()	35 W
()	4.5	8 MB	2.8	65 W
()	35 W

Mobile processors

Ryzen 2000 series

Raven Ridge (Zen/GCN5 based)

Common features of Ryzen 2000 notebook APUs:

Socket: FP5.
The U-series CPUs support DDR4-2400 in dual-channel mode, while the H-series CPUs support it at DDR4-3200 speeds.
L1 cache: 96KB (32KB data + 64KB instruction) per core.
L2 cache: 512KB per core.
All the CPUs support 12 PCIe 3.0 lanes.
Includes integrated GCN 5th generation GPU.
Fabrication process: GlobalFoundries 14LP.

Branding and Model	CPU	GPU	TDP	Release date
Cores (threads)	Clock rate (GHz)	L3 cache (total)	Core config	Model	Clock (GHz)	Config	Processing power (GFLOPS)
Base	Boost
Ryzen 7		4 (8)	3.3	3.8	4MB	1 × 4	RXVega11	1.3	704:44:16 11 CUs	1830.4	35–54W	Sep 10, 2018
	2.2	RXVega10	640:40:16 10 CUs	1664	12–25W	Oct 26, 2017
Ryzen 5		3.2	3.6	Vega8	1.1	512:32:16 8 CUs	1126.4	35–54W	Sep 10, 2018
	2.0	12–25W	Oct 26, 2017
Ryzen 3		4 (4)	3.4	Vega6	384:24:8 6 CUs	844.8	Jan 8, 2018
	2 (4)	2.5	1 × 2	Vega3	192:12:4 3 CUs	422.4

Ryzen 3000 series

Dalí (Zen/GCN5 based)

Common features of Ryzen 3000 notebook APUs:

Socket: FP5.
All the CPUs support DDR4-2400 in dual-channel mode.
L1 cache: 96KB (32KB data + 64KB instruction) per core.
L2 cache: 512KB per core.
All these CPUs provide 12 PCIe 3.0 lanes. 4 are reserved as link to the chipset.
Includes integrated GCN 5th generation GPU.
Fabrication process: GlobalFoundries 14LP.

Branding and Model	CPU	GPU	TDP	Release date
Cores (threads)	Clock rate (GHz)	L3 cache (total)	Core config	Model	Clock (GHz)	Config	Processing power (GFLOPS)
Base	Boost
Ryzen 3		2 (4)	2.6	3.5	4 MB	1 × 2	Vega 3	1.2	192:12:4 3 CUs	460.8	12–25W	Jan 6, 2020
	Sep 22, 2020
	Jan 6, 2019

Picasso (Zen+/GCN5 based)

Common features of Ryzen 3000 notebook APUs:

Socket: FP5.
All the CPUs support DDR4-2400 in dual-channel mode.
L1 cache: 96KB (32KB data + 64KB instruction) per core.
L2 cache: 512KB per core.
All the CPUs support 16 PCIe 3.0 lanes.
Includes integrated GCN 5th generation GPU.
Fabrication process: GlobalFoundries 12LP (14LP+).

Branding and Model	CPU	GPU	TDP	Release date
Cores (threads)	Clock rate (GHz)	L3 cache (total)	Core config	Model	Clock (GHz)	Config	Processing power (GFLOPS)
Base	Boost
Ryzen 7	3780U	4 (8)	2.3	4.0	4 MB	1 × 4	RX Vega 11	1.4	704:44:16 11 CU	1971.2	15 W	Oct 2019
3750H	RX Vega 10	640:40:16 10 CU	1792.0	35 W	Jan 6, 2019
3700C	15 W	Sep 22, 2020
3700U	Jan 6, 2019
Ryzen 5	3580U	2.1	3.7	Vega 9	1.3	576:36:16 9 CU	1497.6	Oct 2019
3550H	Vega 8	1.2	512:32:8 8 CU	1228.8	35 W	Jan 6, 2019
3500C	15 W	Sep 22, 2020
3500U	Jan 6, 2019
3450U	3.5	Jun 2020
Ryzen 3	3350U	4 (4)	Vega 6	384:24:8 6 CU	921.6	Jan 6, 2019
3300U

Ryzen 4000 series

Renoir (Zen 2/GCN5 based)

Common features of Ryzen 4000 notebook APUs:

Socket: FP6.
All the CPUs support DDR4-3200 or LPDDR4-4266 in dual-channel mode.
L1 cache: 64KB (32KB data + 32KB instruction) per core.
L2 cache: 512KB per core.
All the CPUs support 16 PCIe 3.0 lanes.
Includes integrated GCN 5th generation GPU.
Fabrication process: TSMC 7FF.

Branding and Model	CPU	GPU	TDP	Release date
Cores (threads)	Clock rate (GHz)	L3 cache (total)	Core config	Model	Clock (GHz)	Config	Processing power (GFLOPS)
Base	Boost
Ryzen 9		8 (16)	3.3	4.4	8 MB	2 × 4	Radeon Graphics	1.75	512:32:8 8 CU	1792	35–54W	Mar 16, 2020
	3.0	4.3	35W
Ryzen 7		2.9	4.2	1.6	448:28:8 7 CU	1433.6	35–54W
	35W
	2.0	4.4	1.95	512:32:8 8 CU	1996.8	10–25W	Apr 13, 2021
	1.8	4.2	1.75	1792	Mar 16, 2020
	8 (8)	2.0	4.1	1.6	448:28:8 7 CU	1433.6
Ryzen 5		6 (12)	3.0	4.0	2 × 3	1.5	384:24:8 6 CU	1152	35–54W
4600HS	35W
	2.1	448:28:8 7 CU	1344	10–25W	Apr 13, 2021
	384:24:8 6 CU	1152	Mar 16, 2020
	6 (6)	2.3
Ryzen 3		4 (4)	2.7	3.7	4MB	1 × 4	1.4	320:20:8 5 CU	896

Ryzen 5000 series

Lucienne (Zen 2/GCN5 based)

Common features of Ryzen 5000 notebook APUs:

Socket: FP6.
All the CPUs support DDR4-3200 or LPDDR4-4266 in dual-channel mode.
L1 cache: 64KB (32KB data + 32KB instruction) per core.
L2 cache: 512KB per core.
All the CPUs support 16 PCIe 3.0 lanes.
Includes integrated GCN 5th generation GPU.
Fabrication process: TSMC 7FF.

Branding and Model	CPU	GPU	TDP	Release date
Cores (threads)	Clock rate (GHz)	L3 cache (total)	Core config	Model	Clock (GHz)	Config	Processing power (GFLOPS)
Base	Boost
Ryzen 7		8 (16)	1.8	4.3	8MB	2 × 4	Radeon Graphics	1.9	512:32:8 8 CU	1945.6	10–25W	Jan 12, 2021
Ryzen 5		6 (12)	2.1	4.0	2 × 3	1.8	448:28:8 7 CU	1612.8
Ryzen 3		4 (8)	2.6	3.8	4MB	1 × 4	1.5	384:24:8 6 CU	1152

Cezanne and Barceló (Zen 3/GCN5 based)

Cezanne (2021 models), Barceló (2022 models). Common features of Ryzen 5000 notebook APUs:

Socket: FP6.
All the CPUs support DDR4-3200 or LPDDR4-4266 in dual-channel mode.
L1 cache: 64KB (32KB data + 32KB instruction) per core.
L2 cache: 512KB per core.
All the CPUs support 16 PCIe 3.0 lanes.
Includes integrated GCN 5th generation GPU.
Fabrication process: TSMC 7FF.

Branding and model	CPU	GPU	TDP	Release date
Cores (threads)	Clock rate (GHz)	L3 cache (total)	Core config	Model	Clock (GHz)	Config	Processing power (GFLOPS)
Base	Boost (Single Core)	Boost (All Core)
Ryzen 9		8 (16)	3.3	4.8	4.4	16MB	1 × 8	Radeon Graphics	2.1	512:32:8 8 CUs	2150.4	35–54W	Jan 12, 2021
	3.0	4.0	35W
	3.3	4.6	4.2	35–54W
	3.0	4.0	35W
Ryzen 7		3.2	4.4		2.0	2048	35–54W
	2.8	35W
	2.0	4.5	15W	Jan 4, 2022
	1.9	4.4	3.4	10–25W	Jan 12, 2021
Ryzen 5		6 (12)	3.3	4.2		1 × 6	1.8	448:28:8 7 CUs	1612.8	35–54W
	3.0	35W
	2.3	4.3	3.6	15W	Jan 4, 2022
	4.2	10–25W	Jan 12, 2021
	4.0		8MB	1.6	384:24:8 6 CUs	1228.8
	4 (8)	3.3	4.2	1 × 4	1.8	1382.4	35–54W	Jun 23, 2023
Ryzen 3		2.3	4.3	3.8	1.6	1228.8	15W	Jan 4, 2022
	2.7	4.1	10–25W	Jan 12, 2021
	2 (4)	3.0	—N/a	—N/a	1 × 2	1.2	192:12:8 3 CUs	460.8	15W	May 5, 2022

Ryzen 6000 series

Rembrandt (Zen 3+/RDNA2 based)

Common features of Ryzen 6000 notebook APUs:

Socket: FP7, FP7r2.
All the CPUs support DDR5-4800 or LPDDR5-6400 in dual-channel mode.
L1 cache: 64KB (32KB data + 32KB instruction) per core.
L2 cache: 512KB per core.
All the CPUs support 16 PCIe 4.0 lanes.
Native USB 40Gbps (USB4) Ports: 2
Native USB 10Gbps (USB 3.2 Gen 2 operation mode) Ports: 2

Includes integrated RDNA 2 GPU.
Fabrication process: TSMC N6 FinFET.

Branding and model	CPU	GPU	TDP	Release date
Cores (threads)	Clock (GHz)	L3 cache (total)	Core config	Model	Clock (GHz)	Config	Processing power (GFLOPS)
Base	Boost
Ryzen 9		8 (16)	3.3	5.0	16MB	1 × 8	680M	2.4	768:48:8 12 CUs	3686	45W	Jan 4, 2022
	35W
	4.9	45W
	35W
Ryzen 7		3.2	4.7	2.2	3379	45W
	35W
	2.7	15–28W
Ryzen 5		6 (12)	3.3	4.5	1 × 6	660M	1.9	384:24:8 6 CUs	1459	45W
	35W
	2.9	15–28W

Ryzen 7000 series

Mendocino (7020 series, Zen 2/RDNA2 based)

Common features of Ryzen 7020 notebook APUs:

Socket: FT6
All the CPUs support LPDDR5-5500 in dual-channel mode.
L1 cache: 64KB (32KB data + 32KB instruction) per core.
L2 cache: 512KB per core.
All the CPUs support 4 PCIe 3.0 lanes.
Native USB 40Gbps (USB4) Ports: 0
Native USB 10Gbps (USB 3.2 Gen 2 operation mode) Ports: 1
Includes integrated RDNA 2 GPU.
Fabrication process: TSMC N6 FinFET.

Branding and Model	CPU	GPU	TDP	Release date
Cores (threads)	Clock rate (GHz)	L3 cache (total)	Core config	Model	Clock (GHz)	Processing power (GFLOPS)
Base	Boost
Ryzen 5		4 (8)	2.8	4.3	4MB	1 × 4	610M 2 CU	1.9	486	15W	September 20, 2022
Ryzen 3		2.4	4.1

Barcelo-R (7030 series, Zen 3/GCN5 based)

Common features of Ryzen 7030 notebook APUs:

Socket: FP6.
All the CPUs support DDR4-3200 or LPDDR4-4266 in dual-channel mode.
L1 cache: 64KB (32KB data + 32KB instruction) per core.
L2 cache: 512KB per core.
All the CPUs support 16 PCIe 3.0 lanes.
Native USB 40Gbps (USB4) Ports: 0
Native USB 10Gbps (USB 3.2 Gen 2 operation mode) Ports: 2
Includes integrated GCN 5th generation GPU.
Fabrication process: TSMC N7 FinFET.

Branding and Model	CPU	GPU	TDP	Release date
Cores (threads)	Clock rate (GHz)	L3 cache (total)	Core config	Model	Clock (GHz)	Processing power (GFLOPS)
Base	Boost
Ryzen 7	()	8 (16)	2.0	4.5	16MB	1 × 8	Vega 8 CU	2.0	2048	15W	January 4, 2023
Ryzen 5	()	6 (12)	1 × 6	Vega 7 CU	1792
	2.3	4.3	1.8	1613	Q4 2023[citation needed]
Ryzen 3	()	4 (8)	8MB	1 × 4	Vega 6 CU	1382	January 4, 2023

Rembrandt-R (7035 series, Zen 3+/RDNA2 based)

Common features of Ryzen 7035 notebook APUs:

Socket: FP7, FP7r2.
All the CPUs support DDR5-4800 or LPDDR5-6400 in dual-channel mode.
L1 cache: 64KB (32KB data + 32KB instruction) per core.
L2 cache: 512KB per core.
All the CPUs support 16 PCIe 4.0 lanes.
Native USB 40Gbps (USB4) Ports: 0
Native USB 10Gbps (USB 3.2 Gen 2 operation mode) Ports: 2
Includes integrated RDNA 2 GPU.
Fabrication process: TSMC N6 FinFET.

Branding and model	CPU	GPU	TDP	Release date
Cores (threads)	Clock (GHz)	L3 cache (total)	Core config	Model	Clock (GHz)	Processing power (GFLOPS)
Base	Boost
Ryzen 7		8 (16)	3.2	4.75	16MB	1 × 8	680M 12 CU	2.2	3379	35–54W	April 30, 2023

	2.7	4.7	15–28W	January 4, 2023
	4.75	15–30W
	3.1	4.5	—N/a	35–54W	2024

Ryzen 5		6 (12)	3.3	4.55	1 × 6	660M 6 CU	1.9	1459	April 30, 2023

	2.9	15–30W	January 4, 2023
	4 (8)	3.2	4.2	8MB	1 × 4	—N/a	35–53W	2024

Ryzen 3		3.0	4.3	660M 4 CU	1.8	922	15–30W	January 4, 2023

Phoenix (7040 series, Zen 4/RDNA3/XDNA based)

Common features of Ryzen 7040 notebook APUs:

Socket: FP7, FP7r2, FP8.
All the CPUs support DDR5-5600 or LPDDR5X-7500 in dual-channel mode.
L1 cache: 64KB (32KB data + 32KB instruction) per core.
L2 cache: 1MB per core.
All models support AVX-512 using a half-width 256-bit FPU.
PCIe 4.0 support.
Native USB 40Gbps (USB4) Ports: 2
Native USB 10Gbps (USB 3.2 Gen 2 operation mode) Ports: 2
Includes integrated RDNA 3 GPU.
Some models include XDNA AI Engine (Ryzen AI) NPU capable of 10 TOPS.
Fabrication process: TSMC N4 FinFET.

Branding and model	CPU	GPU	NPU (Ryzen AI)	TDP	Release date
Cores (threads)	Clock (GHz)	L3 cache (total)	Core config	Model	Clock (GHz)
Total	Zen 4	Zen 4c	Base	Boost
Ryzen 9	()	8 (16)	8 (16)	—N/a	4.0	5.2	16 MB	1 × 8	780M 12 CU	2.8	Yes	35-54 W	April 30, 2023
7940H
Ryzen 7	()	3.8	5.1	2.7
7840H
()	3.3	15-30 W	May 3, 2023
	6 (12)	2 (4)	4 (8)	3.2	4.7	2 + 4	740M 4 CU	No	20-40 W	May 3, 2023
Ryzen 5	()	6 (12)	6 (12)	—N/a	4.3	5.0	1 × 6	760M 8 CU	2.6	Yes	35-54 W	April 30, 2023
7640H
()	3.5	4.9	15-30 W	May 3, 2023
()	2 (4)	4 (8)	3.7 / 3.0	4.9 / 3.5	2 + 4	740M 4 CU	2.8	No	November 2, 2023
()	6 (12)	—N/a	3.2	4.9	1 × 6	2.5	May 3, 2023
Ryzen 3		4 (8)	1 (2)	3 (6)	3.6 / 2.8	4.7 / 3.3	8 MB	1 + 3

Dragon Range (7045 series, Zen 4/RDNA2 based)

Common features of Ryzen 7045 notebook CPUs:

Socket: FL1.
All the CPUs support DDR5-5200 in dual-channel mode.
L1 cache: 64KB (32KB data + 32KB instruction) per core.
L2 cache: 1MB per core.
All the CPUs support 28 PCIe 5.0 lanes.
Native USB 40Gbps (USB4) Ports: 0
Native USB 10Gbps (USB 3.2 Gen 2 operation mode) Ports: 4
Includes integrated RDNA 2 GPU on the I/O die with 2 CUs and clock speeds of 400MHz (base), 2.2GHz (boost).
Fabrication process: TSMC N5 FinFET (N6 FinFET for the I/O and graphics die).

Branding and model	Cores (threads)	Clock (GHz)	L3 cache (total)	Chiplets	Core config	TDP	Release date
Base	Boost
Ryzen 9		16 (32)	2.3	5.4	128MB	2 × CCD 1 × I/OD	2 × 8	55–75W	July 27, 2023
	2.5	64MB	February 28, 2023
	2.4	5.2	January 17, 2024
	12 (24)	3.0	5.2	2 × 6	45–75W	February 28, 2023
Ryzen 7		2.9	5.1	January 17, 2024
	8 (16)	3.6	5.1	32MB	1 × CCD 1 × I/OD	1 × 8	February 28, 2023
Ryzen 5		6 (12)	4.0	5.0	1 × 6

Ryzen 8000 series

Hawk Point (8040 series, Zen 4/RDNA3/XDNA based)

Key features of Ryzen 8040 notebook APUs:

Socket: BGA (FP7, FP7r2 or FP8 type packages).
All models support DDR5-5600 or LPDDR5X-7500 in dual-channel mode.
CPU uses Zen4 cores (Phoenix) or a combination of Zen4 and Zen4c cores (Phoenix2).
GPU uses the RDNA 3 (Navi 3) architecture.
Some models include first-generation Ryzen AI NPU (XDNA) capable of 16 TOPS.
All models support AVX-512 using a half-width 256-bit FPU.
PCIe 4.0 support.
Native USB 40Gbps (USB4) Ports: 2
Native USB 10Gbps (USB 3.2 Gen 2 operation mode) Ports: 2
Fabrication process: TSMC N4 FinFET.

Branding and model	CPU	GPU	NPU (Ryzen AI)	TDP	Release date
Cores (threads)	Clock (GHz)	L3 cache (total)	Core config	Model	Clock (GHz)
Total	Zen 4	Zen 4c	Base	Boost
Ryzen 9		8 (16)	8 (16)	—	4.0	5.2	16 MB	1 × 8	780M 12 CU	2.8	Yes	35-54 W	December 6, 2023
Ryzen 7		3.8	5.1	2.7
	3.3	20-30 W
	15-30 W
8745HS	3.8	4.9	2.6	No	35–54 W	?
8745H
Ryzen 5		6 (12)	6 (12)	4.3	5.0	1 × 6	760M 8 CU	2.6	Yes	35-54 W	December 6, 2023
	3.5	4.9	20-30 W
	15-30 W
	2 (4)	4 (8)	3.7 / 3.0	4.9 / 3.5	2 + 4	740M 4 CU	2.8	No	?
Ryzen 3		4 (8)	1 (2)	3 (6)	3.6 / 2.8	4.7 / 3.3	8 MB	1 + 3	2.5

Dragon Range Refresh (8045 series, Zen4/RDNA2 based)

Common features of Ryzen 8045 notebook CPUs:

Socket: FL1.
All the CPUs support DDR5-5200 in dual-channel mode.
L1 cache: 64KB (32KB data + 32KB instruction) per core.
L2 cache: 1MB per core.
All the CPUs support 28 PCIe 5.0 lanes.
Native USB 40Gbps (USB4) Ports: 0
Native USB 10Gbps (USB 3.2 Gen 2 operation mode) Ports: 4
Includes integrated RDNA 2 GPU on the I/O die with 2 CUs and clock speeds of 400MHz (base), 2.2GHz (boost).
Fabrication process: TSMC N5 FinFET (N6 FinFET for the I/O and graphics die).

Branding and model	Cores (threads)	Clock (GHz)	L3 cache (total)	Chiplets	Core config	TDP	Release date
Base	Boost
Ryzen 9		16 (32)	2.5	5.4	64MB	2 × CCD 1 × I/OD	2 × 8	55–75W	April 23, 2025
	2.4	5.3
Ryzen 7		12 (24)	2.9	5.1	2 × 6	45–75W
	8 (16)	3.6	32MB	1 × CCD 1 × I/OD	1 × 8

Ryzen 9000 series

Fire Range (Zen 5/RDNA2 based)

Common features of Ryzen 9000 Fire Range series:

Socket: FL1
All Models support DDR5-5600 in 128-bit mode; Maximum memory: 96 GB
All models support 28 PCIe 5.0 lanes
Native USB 10Gbps (USB 3.2 Gen 2 operation mode): 4
Native USB 2.0 (480 Mbit/s): 1
iGPU: AMD Radeon 610M (2 CU @ 2200MHz)
No NPU
L1 cache: 80 KB (48 KB data + 32 KB instruction) per core
L2 cache: 1 MB per core
Fabrication process: TSMC N4 FinFET (CCD) + TSMC N6 FinFET (I/OD)

Branding and Model	Cores (threads)	Clock (GHz)	L3 cache (total)	Chiplets	Core config	TDP	Release date
Base	Boost
Ryzen 9		16 (32)	2.3	5.4	128MB	2 × CCD 1 × I/OD	2 × 8	55–75W	1H 2025
	2.5	64MB
	12 (24)	3.0	5.2	2 × 6	45–75W

Ryzen 10 Series

Mendocino (10 series, Zen 2/RDNA2 based)

Common features of Ryzen 10 notebook APUs:

Socket: FT6
All the CPUs support LPDDR5-5500 in dual-channel mode.
L1 cache: 64KB (32KB data + 32KB instruction) per core.
L2 cache: 512KB per core.
All the CPUs support 4 PCIe 3.0 lanes.
Native USB 40Gbps (USB4) Ports: 0
Native USB 10Gbps (USB 3.2 Gen 2 operation mode) Ports: 1
Includes integrated RDNA 2 GPU.
Fabrication process: TSMC N6 FinFET.

Branding and Model	CPU	GPU	TDP	Release date
Cores (threads)	Clock rate (GHz)	L3 cache (total)	Core config	Model	Clock (GHz)	Processing power (GFLOPS)
Base	Boost
Ryzen 5		4 (8)	2.8	4.3	4MB	1 × 4	610M 2 CU	1.9	486	15W	October 1, 2025
Ryzen 3		2.4	4.1

Ryzen 100 series

Rembrandt-R (100 series, Zen 3+/RDNA2 based)

Common features of Ryzen 100 notebook APUs:

Socket: FP7, FP7r2.
All the CPUs support DDR5-4800 or LPDDR5-6400 in dual-channel mode.
L1 cache: 64KB (32KB data + 32KB instruction) per core.
L2 cache: 512KB per core.
All the CPUs support 16 PCIe 4.0 lanes.
Native USB 40Gbps (USB4) Ports: 0
Native USB 10Gbps (USB 3.2 Gen 2 operation mode) Ports: 2
Includes integrated RDNA 2 GPU.
Fabrication process: TSMC N6 FinFET.

Branding and model	CPU	GPU	TDP	Release date
Cores (threads)	Clock (GHz)	L3 cache (total)	Core config	Model	Clock (GHz)	Processing power (GFLOPS)
Base	Boost
Ryzen 7		8 (16)	3.2	4.75	16MB	1 × 8	680M 12 CU	2.2	3379	35–54W	October 1, 2025
	2.7	15–30W
Ryzen 5		6 (12)	3.3	4.55	1 × 6	660M 6 CU	1.9	1459	35–54W
	2.9	15–30W
Ryzen 3		4 (8)	3.0	4.3	8MB	1 × 4	660M 4 CU	1.8	922

Ryzen 200 series

Hawk Point Refresh (200 series, Zen 4/RDNA3/XDNA based)

Branding and Model	CPU	GPU	Ryzen AI	TDP	Release date
Cores (threads)	Clock (GHz)	L3 cache (total)	Core config	Model	Clock (GHz)	Processing power
Total	Zen 4	Zen 4c	Base	Boost
Single （GFLOPS）
Ryzen 9		8 (16)	8 (16)	—N/a	4.0	5.2	16 MB	1 × 8	780M 12 CU	2.8	4357	Yes	35–54 W	February 18, 2025
Ryzen 7		3.8	5.1	2.7	4201
()	3.3	15–30 W
Ryzen 5		6 (12)	6 (12)	4.3	5.0	1 × 6	760M 8 CU	2.6	2663	35–54 W
()	3.5	4.9	15–30 W
()	2 (4)	4 (8)	3.7 / 3.0	4.9 / 3.5	2 + 4	740M 4 CU	2.8	1434	No
	3.2 / 3.2	4.7 / 3.5	2.7	1383	March 27, 2025
Ryzen 3	()	4 (8)	1 (2)	3 (6)	3.6 / 2.8	4.7 / 3.3	8 MB	1 + 3	2.5	1280	February 18, 2025

Ryzen AI 300 series

Strix Point and Krackan Point (Zen 5/RDNA3.5/XDNA2 based)

Common features of Ryzen AI 300 notebook APUs:

Socket: BGA, FP8 package type.
All models support DDR5-5600 or LPDDR5X-8000 in dual-channel mode.
All models support 16 PCIe 4.0 lanes.
Native USB 40Gbps (USB4) Ports: 2
Native USB 10Gbps (USB 3.2 Gen 2 operation mode) Ports: 2
iGPU uses the RDNA 3.5 microarchitecture.
NPU uses the XDNA 2 AI Engine (Ryzen AI).
Both Zen 5 and Zen 5c cores support AVX-512 using a half-width 256-bit FPU.
L1 cache: 80KB (48KB data + 32KB instruction) per core.
L2 cache: 1MB per core.
Fabrication process: TSMC N4P FinFET.

Branding and model	CPU	GPU	NPU (Ryzen AI)	TDP	Release date
Cores (threads)	Clock (GHz)	L3 cache (total)	Model	Clock (GHz)
Total	Zen 5	Zen 5c	Base	Boost (Zen 5)	Boost (Zen 5c)
Ryzen AI 9	()	12 (24)	4 (8)	8 (16)	2.0	5.1	3.3	24MB	890M 16 CUs	2.9	55TOPS	15–54W	June 2, 2024
()	50TOPS
	10 (20)	6 (12)	5.0	880M 12 CUs
Ryzen AI 7		8 (16)	3 (6)	5 (10)	16MB	October 10, 2024
()	4 (8)	4 (8)	3.5	860M 8 CU	3.0	February 18, 2025
Ryzen AI 5	()	6 (12)	3 (6)	3 (6)	4.8	3.4	840M 4 CU	2.9
	4 (8)	1 (2)	4.5	8MB	820M 2 CU	2.8	15–28W	July 2025

Strix Halo (Zen 5/RDNA3.5/XDNA2 based)

Common features of Strix Halo mobile CPUs:

Socket: BGA, FP11 package type.
All CPUs only support soldered LPDDR5X memory with a 256-bit memory bus.
All CPUs support 16 lanes of PCIe 4.0 lanes.
iGPU uses the RDNA 3.5 architecture.
Fabrication process: TSMC N4P FinFET.

Branding and Model	CPU	GPU	Memory	NPU (Ryzen AI)	Chiplets	Core config	TDP	Release date
Cores (threads)	Clock (GHz)	L3 cache (total)	Model	Clock (GHz)	Bustype &width (Maxmemory)	Clock (MT/s)	Bandwidth (GB/s)
Base	Boost
Ryzen AI MAX+	()	16 (32)	3.0	5.1	64MB	8060S 40 CUs	2.9	LPDDR5X 256-bit (128 GB)	Up to 8000	Up to 256	50 TOPS	2 × CCD 1 × I/ODwithGPU	2 × 8	45–120W	Q1 2025
	12 (24)	3.2	5.0	2 × 6	Q1 2026
Ryzen AI MAX	()	8050S 32 CUs	2.8	Q1 2025
Ryzen AI MAX+		8 (16)	3.6	32MB	8060S 40 CUs	2.9	1 × CCD 1 × I/ODwithGPU	1 × 8	Q1 2026
Ryzen AI MAX	()	8050S 32 CUs	2.8	Q1 2025
	6 (12)	4.9	16MB	8040S 16 CUs	LPDDR5X 128-bit (64 GB)	Up to 128	1 × 6

Ryzen AI 400 series

Gorgon Point (Zen 5/RDNA3.5/XDNA2 based)

Common features of Ryzen AI 400 notebook APUs:

Socket: BGA, FP8 package type.
All models support DDR5-5600 or LPDDR5X-8000 (LPDDR5X-8533 for 440, 450, 465, 470, 475 series) in dual-channel mode.
All models support 16 PCIe 4.0 lanes.
Native USB 40Gbps (USB4) Ports: 2
Native USB 10Gbps (USB 3.2 Gen 2 operation mode) Ports: 2
iGPU uses the RDNA 3.5 microarchitecture.
NPU uses the XDNA 2 AI Engine (Ryzen AI).
Both Zen 5 and Zen 5c cores support AVX-512 using a half-width 256-bit FPU.
L1 cache: 80KB (48KB data + 32KB instruction) per core.
L2 cache: 1MB per core.
Fabrication process: TSMC N4P FinFET.

Branding and model	CPU	GPU	NPU (Ryzen AI)	TDP	Release date
Cores (threads)	Clock (GHz)	L3 cache (total)	Model	Clock (GHz)
Total	Zen 5	Zen 5c	Base	Boost (Zen 5)	Boost (Zen 5c)
Ryzen AI 9	()	12 (24)	4 (8)	8 (16)	2.0	5.2	3.3	24MB	890M 16 CUs	3.1	60TOPS	15–54W	January 5, 2026
()	55TOPS
()	10 (20)	6 (12)	5.0	880M 12 CUs	2.9	50TOPS
Ryzen AI 7	()	8 (16)	4 (8)	4 (8)	5.1	3.6	16MB	860M 8 CUs	3.1
	6 (12)	2 (4)	4.6	3.5	8MB	840M 4 CUs	2.9
Ryzen AI 5		3 (6)	3 (6)	4.8	16MB
()	2 (4)	4 (8)	4.5	3.4	8MB	2.8
	4 (8)	1 (2)	3 (6)	15–28W

Gorgon Halo (Zen 5/RDNA3.5/XDNA2 based)

Common features of Ryzen AI Max PRO 400 notebook APUs:

Socket: BGA, FP11 package type.
All CPUs only support soldered LPDDR5X memory with a 256-bit memory bus.
All CPUs support 16 PCIe 4.0 lanes.
Native USB 40Gbps (USB4) Ports: 2
Native USB 10Gbps (USB 3.2 Gen 2 operation mode) Ports: 3
NVMe Support: Boot, RAID0, RAID1
iGPU uses the RDNA 3.5 microarchitecture.
NPU uses the XDNA 2 AI Engine (Ryzen AI).
Zen 5 cores support AVX-512 using a half-width 256-bit FPU.
L2 cache: 1 MB per core.
Fabrication process: TSMC N4P FinFET.

Branding and Model	CPU	GPU	Memory	NPU (Ryzen AI)	Chiplets	Core config	TDP	Release date
Cores (threads)	Clock (GHz)	L3 cache (total)	Model	Clock (GHz)	Bus type & width (Max memory)	Clock (MT/s)	Bandwidth (GB/s)
Base	Boost
Ryzen AI MAX+		16 (32)	3.1	5.2	64 MB	8065S 40 CUs	3.0	LPDDR5X 256-bit (192 GB)	Up to 8533	Up to 273	55 TOPS	2 × CCD 1 × I/ODwithGPU	2 × 8	55W cTDP: 45–120W	Q3 2026
Ryzen AI MAX		12 (24)	3.2	5.0	8050S 32 CUs	2.8	50 TOPS	2 × 6
	8 (16)	3.6	32 MB	1 × CCD 1 × I/ODwithGPU	1 × 8

Handheld gaming PC processors

Ryzen Z1 series (Zen 4/RDNA3 based)

Common features of Ryzen Z1 handheld APUs:

Socket: FP7, FP7r2, FP8.
All the CPUs support DDR5-5600 or LPDDR5X-7500 in dual-channel mode.
L1 cache: 64KB (32KB data + 32KB instruction) per core.
L2 cache: 1MB per core.
All the CPUs support 20 PCIe 4.0 lanes.
Includes integrated RDNA 3 GPU.
No NPU.
Fabrication process: TSMC 4 nm FinFET.

Branding and model	CPU	GPU	Default TDP	Configurable TDP	Release date
Cores (threads)	Clock (GHz)	L3 cache (total)	Core config	Model	Clock (GHz)
Total	Zen 4	Zen 4c	Base	Boost
Ryzen		8 (16)	8 (16)	—N/a	3.3	5.1	16MB	1 × 8	RDNA 3 12 CU	2.9	28W	9–30W	May 2023
	6 (12)	2 (4)	4 (8)	3.2 (average) 3.7 (Zen 4) 3 (Zen 4c)	4.9 (Zen 4) 3.5 (Zen 4c)	2 + 4	RDNA 3 4 CU	2.8

Ryzen Z2 series (multiple architectures)

Branding and Model	Codename	Process	CPU	GPU	NPU (Ryzen AI)	Default TDP	Configurable TDP	Release date
Cores (threads)	Clock (GHz)	L3 cache (total)	Core config	Architecture	Compute units	Clock (GHz)
Total	Zen 2/3+/4/5	Zen 5c	Base	Boost
Ryzen AI		Strix Point	TSMC 4 nm	8 (16)	3 (6) (Zen 5)	5 (10) (Zen 5c)	2.0	5.0 / 3.3	16MB	3 + 5	RDNA3.5	16	2.9	50TOPS	28W	15–35W	June 2025
Ryzen		—N/a	Q1 2025
	Hawk Point	TSMC 4 nm	8 (16) (Zen 4)	—N/a	3.3	5.1	1 × 8	RDNA3	12	2.7	15–30W
	Rembrandt+	TSMC 6 nm	4 (8)	4 (8) (Zen 3+)	3.0	4.3	8MB	1 × 4	RDNA2	12	2.2
	Van Gogh	TSMC 7 nm	4 (8) (Zen 2)	2.8	3.8	4MB	8	1.8	15W	6–20W	June 2025

Embedded processors

1000 series

Great Horned Owl (V1000 series, Zen/GCN5 based)

Model	Release date	Fab	CPU	GPU	Memory support	TDP	Junction temp. range (°C)
Cores (threads)	Clock rate (GHz)	Cache	Model	Config	Clock (GHz)	Processing power (GFLOPS)
Base	Boost	L1	L2	L3
	February 2018	GloFo 14LP	2 (4)	2.3	3.2	64KB inst. 32KB data per core	512 KB per core	4 MB	Vega 3	192:12:16 3 CU	1.0	384	DDR4-2400 dual-channel	12–25W	0–105
	December 2018	4 (8)	2.0	3.6	Vega 8	512:32:16 8 CU	1.1	1126.4	-40–105
	2.2	—N/a	—N/a	0–105
	February 2018	2.0	3.6	Vega 8	512:32:16 8 CU	1.1	1126.4
	3.25	DDR4-3200 dual-channel	35–54W
	December 2018	3.35	—N/a
	February 2018	3.8	Vega 11	704:44:16 11 CU	1.3	1830.4

Banded Kestrel (R1000 series, Zen/GCN5 based)

Model	Release date	Fab	CPU	GPU	Memory support	TDP
Cores (threads)	Clock rate (GHz)	Cache	Model	Config	Clock (GHz)	Processing power (GFLOPS)
Base	Boost	L1	L2	L3
	February 25, 2020	GloFo 14LP	2 (2)	1.2	2.6	64 KB inst. 32 KB data per core	512 KB per core	4 MB	Vega 3	192:12:4 3 CU	1.0	384	DDR4-2400 single-channel	6W
	2 (4)	1.5	2.8	DDR4-2400 dual-channel	8-10W
	April 16, 2019	2.4	3.3	12–25W
	2.6	3.5	1.2	460.8

2000 series

Grey Hawk (V2000 series, Zen 2/GCN5 based)

Model	Release date	Fab	CPU	GPU	Socket	PCIe support	Memory support	TDP
Cores (threads)	Clock rate (GHz)	Cache	Archi- tecture	Config	Clock (GHz)	Processing power (GFLOPS)
Base	Boost	L1	L2	L3
	Nov 10, 2020	TSMC 7FF	6 (12)	2.1	3.95	32 KB inst. 32 KB data per core	512KB per core	8MB	GCN 5	384:24:8 6 CU	1.5	1152	FP6	20 (8+4+4+4) PCIe 3.0	DDR4-3200dual-channel LPDDR4X-4266quad-channel	10–25W
	3.0	3.95	35–54W
	Jan 4, 2023	3.2	448:28:8 7 CU	1.6	1433.6
	Nov 10, 2020	8 (16)	1.7	4.15	10–25W
	2.9	4.25	35–54W

River Hawk (R2000 series, Zen+ based)

Model	Release date	Fab	CPU	GPU	Socket	PCIe support	Memory support	TDP
Cores (threads)	Clock rate (GHz)	Cache	Archi- tecture	Config	Clock (GHz)	Processing power (GFLOPS)
Base	Boost	L1	L2	L3
	June 7, 2022	GloFo 12LP	2 (4)	2.7	3.5	64KB inst. 32KB data per core	512KB per core	4MB	GCN 5	192:12:4 3 CU	1.2	460.8	FP5	8 lanes Gen 3	DDR4-2400dual-channel ECC	10–25W
	4 (4)	2.1	384:24:8 6 CU	921.6	16 lanes Gen 3	DDR4-2666dual-channel ECC	10–35W

3000 series

(V3000 series, Zen 3 based)

Model	Release date	Fab	CPU	Socket	PCIe support	Memory support	TDP
Cores (threads)	Clock rate (GHz)	Cache
Base	Boost	L1	L2	L3
V3C14	September 27, 2022	TSMC 7FF	4 (8)	2.3	3.8	32 KB inst. 32 KB data per core	512KB per core	8MB	FP7r2	20 (8+4+4+4) PCIe 4.0	DDR5-4800dual-channel	15W
V3C44	3.5	3.8	45W
V3C16	6 (12)	2.0	3.8	16MB	15W
V3C18I	8 (16)	1.9	3.8	15W
V3C48	3.3	3.8	45W

5000 series

(Ryzen Embedded 5000 series, Zen 3 based)

Common features of Ryzen Embedded 5000 series CPUs:

Socket: AM4.
All the CPUs support ECC DDR4-3200 dual-channel mode.
L1 cache: 64KB (32KB data + 32KB instruction) per core.
L2 cache: 512KB per core.
No integrated graphics.
All the CPUs support 24 PCIe 4.0 lanes. 4 of the lanes are reserved as link to the chipset.
Fabrication process: TSMC 7FF.

Processor branding	Model	Cores (threads)	Clock rate (GHz)	L3 cache (total)	Chiplets	Core config	TDP	Release date
Base	Boost
Ryzen Embedded		16 (32)	3.05	3.4	64MB	2 × CCD 1 × I/OD	2 × 8	105W	Apr 20, 2023
	12 (24)	3.35	3.7	2 × 6
	8 (16)	3.4	32MB	1 × CCD 1 × I/OD	1 × 8	65-100W
	6 (12)	3.3	3.6	1 × 6	65W

7000 series

(Ryzen Embedded 7000 series, Zen 4/RDNA2 based)

Common features of Ryzen Embedded 7000 series CPUs:

Socket: AM5.
All the CPUs support ECC DDR5-5200 dual-channel mode.
L1 cache: 64KB (32KB data + 32KB instruction) per core.
L2 cache: 1MB per core.
All the CPUs support 28 PCIe 5.0 lanes. 4 of the lanes are reserved as link to the chipset.
Includes integrated RDNA 2 GPU with 2 CUs and boost clock speeds of 2200 MHz.
Fabrication process: TSMC N5 FinFET (N6 FinFET for the I/O die).

Processor branding	Model	Cores (threads)	Clock rate (GHz)	L3 cache (total)	Chiplets	Core config	TDP	Release date
Base	Boost
Ryzen Embedded		12 (24)	3.7	5.4	64MB	2 × CCD 1 × I/OD	2 × 6	65W	Nov 14, 2023
	8 (16)	3.8	5.3	32MB	1 × CCD 1 × I/OD	1 × 8
	4.5	5.4	105W
	6 (12)	3.8	5.1	1 × 6	65W
	4.7	5.3	105W

P100 series

(Ryzen AI Embedded P100 series, Zen 5/RDNA 3.5 based)

Model numbers with the "i" suffix are marketed for industrial use and support operating at colder ambient temperatures. Model numbers with the "a" suffix are marketed for automotive use and are AEC-Q100 qualified for reliability and longevity.

Common features of Ryzen AI Embedded P100 series CPUs:

Socket: BGA, FP8 package type.
L3 cache (total): 8MB.
All models support dual-channel ECC DRAM.
All models support 14 PCIe 4.0 lanes.
Native 10GbE ports (with TSN): 2
Native USB 5Gbps (USB 3.2 Gen 1 operation mode): 1
Native USB 5Gbps (USB 3.1 Gen 1 operation mode): 1
Native USB 2.0 (480 Mbit/s): 4
iGPU uses the RDNA 3.5 microarchitecture and supports output at up to 120FPS to up to either 4 × 4K or 2 × 8K displays.
iGPU supports encoding/decoding video at up to 4K & 60Hz.
All iGPUs support HDMI 2.1 and DisplayPort 2.0 output.
NPU uses the XDNA 2 architecture.
Rated longevity: "2.5 years (Standard), Up to 10 years (Extended)"

Processor branding	Model	CPU	GPU	Memory support	NPU TOPS	USB4 ports	Nominal TDP	Configurable TDP	Junction temperature range	Release date
Cores (threads)	Boost clock (GHz)	L2 cache (total)	CU	Clock (GHz)	eDP version
Ryzen AI Embedded		6 (12)	4.5	6MB	4	2.8	1.4	LPDDR5X-8000 DDR5-5600	50	2	28W	15–54W	−40°C–105°C	Q2 2026
	0°C–105°C
	3.7	4MB	2.4	1.5	LPDDR5X-7500 with RAS	No	45W	25–45W	−40°C–105°C
	4 (8)	4.4	4MB	2	2.7	1.4	LPDDR5X-7500 DDR5-5600	30	2	28W	15–54W	−40°C–105°C
	0°C–105°C
	3.7	4	2.0	1.6	LPDDR5X-7500 with RAS	No	15–30W	−40°C–105°C