Latest:NVIDIA H20 AI GPUs Help China Tech Dominate

The nation of China Plans to Begin Mass Producing GPU manufacturer NVIDIA H20 AI GPUs in Q2 2024, Absolutely In accordance With US Laws

[{"selector":"#anim-f96ea9da-e955-483f-9e22-001acd324b0f","keyframes":{"opacity":[0,1]},"delay":0,"duration":600,"easing":"cubic-bezier(0.2, 0.6, 0.0, 1)","fill":"both"}] [{"selector":"#anim-70208fd6-e590-4286-beef-cc7e52d39203","keyframes":{"transform":["translate3d(0px, -137.19577%, 0)","translate3d(0px, 0px, 0)"]},"delay":0,"duration":600,"easing":"cubic-bezier(0.2, 0.6, 0.0, 1)","fill":"both"}] [{"selector":"#anim-7423667f-ab29-4b07-8db4-8cd3d2f6aaa0 [data-leaf-element=\"true\"]","keyframes":{"transform":["translate3d(-34.19732431882569%, 0, 0)","translate3d(0%, 0, 0)"]},"delay":0,"duration":2000,"easing":"cubic-bezier(.3,0,.55,1)","fill":"both"}]

NVIDIA Will Start Shipping AI GPUs to China in Q2 2024, With NVIDIA H20 AI GPUs Mass Production To Begin.

[{"selector":"#anim-39eaf04b-3e41-49f8-979a-fa3867c6dfde","keyframes":{"transform":["translate3d(-126.39405%, 0px, 0)","translate3d(0px, 0px, 0)"]},"delay":0,"duration":600,"easing":"cubic-bezier(0.4, 0.4, 0.0, 1)","fill":"both"}] [{"selector":"#anim-4bab202d-69db-417e-bbe3-0216b714a6d5","keyframes":{"opacity":[0,1]},"delay":0,"duration":600,"easing":"cubic-bezier(0.4, 0.4, 0.0, 1)","fill":"both"}] [{"selector":"#anim-e0eb7103-1be0-4d19-93de-5e224644973e","keyframes":{"transform":["scale(0.15)","scale(1)"]},"delay":0,"duration":600,"easing":"cubic-bezier(0.4, 0.4, 0.0, 1)","fill":"forwards"}] [{"selector":"#anim-cd612bf5-e00b-455b-a5b7-70897e8bc1c8 [data-leaf-element=\"true\"]","keyframes":{"transform":["translate3d(-34.19732431882569%, 0, 0) translate(-25%, 0%) scale(1.5)","translate3d(0%, 0, 0) translate(0%, 0%) scale(1)"]},"delay":0,"duration":2000,"fill":"forwards"}]

November of last year saw reports that NVIDIA was getting ready to release a new line of NVIDIA H20 AI GPUs for the Chinese market in place of the H100, L40, and L4 chips, which were prohibited by US policy

[{"selector":"#anim-5a1c17c4-e943-4822-bf7b-d5a152abd362","keyframes":{"transform":["translate3d(-122.61484%, 0px, 0)","translate3d(0px, 0px, 0)"]},"delay":0,"duration":600,"easing":"cubic-bezier(0.4, 0.4, 0.0, 1)","fill":"both"}] [{"selector":"#anim-036e5491-7749-4d69-9728-14ea9217ba85","keyframes":{"opacity":[0,1]},"delay":0,"duration":600,"easing":"cubic-bezier(0.4, 0.4, 0.0, 1)","fill":"both"}] [{"selector":"#anim-202b7509-07d2-4724-9ab9-827369ee4b91","keyframes":{"transform":["scale(0.15)","scale(1)"]},"delay":0,"duration":600,"easing":"cubic-bezier(0.4, 0.4, 0.0, 1)","fill":"forwards"}] [{"selector":"#anim-56d0cbf5-cf75-408f-93b5-20866bab4676 [data-leaf-element=\"true\"]","keyframes":{"transform":["translate3d(-30.940933950284393%, 0, 0) translate(-25%, 0%) scale(1.5)","translate3d(0%, 0, 0) translate(0%, 0%) scale(1)"]},"delay":0,"duration":2000,"fill":"forwards"}]

Technically speaking, the NVIDIA H20 AI GPUs is a simplified version of the H100 GPU, with 96 GB of RAM that can operate at up to 4.0 Tb/s, 296 TFLOPs of compute capability

[{"selector":"#anim-3c03989e-a7fa-42c9-a4bb-99195d6fcbdc [data-leaf-element=\"true\"]","keyframes":{"transform":["translate3d(-34.249999904429615%, 0, 0)","translate3d(0%, 0, 0)"]},"delay":0,"duration":2000,"easing":"cubic-bezier(.3,0,.55,1)","fill":"both"}] [{"selector":"#anim-803e9a97-a8b1-4d1e-9e65-7e1d3f13a697","keyframes":{"transform":["scale(1)","scale(1.5)","scale(0.95)","scale(1)"],"offset":[0,0.33,0.66,1]},"delay":0,"duration":1450,"easing":"ease-in-out","fill":"both","iterations":1}]

FLOPs are FP16 Tensor Core FLOPs (with Sparsity), not INT8 or FP8 FLOPs. The GPU is an HGX solution with 8-way configurations and a 400W TDP

[{"selector":"#anim-425fc41d-c0a0-4dc2-a2e2-5901b96720e3","keyframes":{"transform":["translate3d(-121.10727%, 0px, 0)","translate3d(0px, 0px, 0)"]},"delay":0,"duration":600,"easing":"cubic-bezier(0.4, 0.4, 0.0, 1)","fill":"both"}] [{"selector":"#anim-ddc1469a-cd8d-4d01-b435-862041b3b00a","keyframes":{"opacity":[0,1]},"delay":0,"duration":600,"easing":"cubic-bezier(0.4, 0.4, 0.0, 1)","fill":"both"}] [{"selector":"#anim-97a5dde6-0f42-4e1d-8dc8-f51ef6d894ea","keyframes":{"transform":["scale(0.15)","scale(1)"]},"delay":0,"duration":600,"easing":"cubic-bezier(0.4, 0.4, 0.0, 1)","fill":"forwards"}] [{"selector":"#anim-aeb94bfe-6c80-4fa7-9a9e-e9ad69e66aa5 [data-leaf-element=\"true\"]","keyframes":{"transform":["translate3d(-34.249999904429615%, 0, 0) translate(-25%, 0%) scale(1.5)","translate3d(0%, 0, 0) translate(0%, 0%) scale(1)"]},"delay":0,"duration":2000,"fill":"forwards"}]

The business is currently anticipating starting mass production of its H20 and the other AI GPUs in Q2 2024. Wistron, a significant substrate supplier for the green team, will manage this process

[{"selector":"#anim-11b0363f-7cc5-472d-89ba-7a33877a1b27","keyframes":{"transform":["translate3d(-120.54795%, 0px, 0)","translate3d(0px, 0px, 0)"]},"delay":0,"duration":1000,"easing":"cubic-bezier(.2, 0, .8, 1)","fill":"both"}] [{"selector":"#anim-ea7ea34c-30db-45d5-86e9-c081810f8515","keyframes":{"transform":["rotateZ(-180deg)","rotateZ(0deg)"]},"delay":0,"duration":1000,"easing":"cubic-bezier(.2, 0, .5, 1)","fill":"forwards"}] [{"selector":"#anim-e1a761e8-93ff-47b8-9a06-5882ef0d79cd [data-leaf-element=\"true\"]","keyframes":{"transform":["translate3d(36.92059262580184%, 0, 0)","translate3d(0%, 0, 0)"]},"delay":0,"duration":2000,"easing":"cubic-bezier(.3,0,.55,1)","fill":"both"}]

Due to constraints on US-based tech businesses, Huawei is likely to earn a significant market share in the nation

[{"selector":"#anim-7cf4d88e-479f-4a70-a4d5-73ac45df0192","keyframes":{"transform":["translate3d(-123.57143%, 0px, 0)","translate3d(0px, 0px, 0)"]},"delay":0,"duration":600,"easing":"cubic-bezier(0.4, 0.4, 0.0, 1)","fill":"both"}] [{"selector":"#anim-ac2cd6da-e4bf-4b22-94b0-896300157d86","keyframes":{"opacity":[0,1]},"delay":0,"duration":600,"easing":"cubic-bezier(0.4, 0.4, 0.0, 1)","fill":"both"}] [{"selector":"#anim-55f5302d-522b-4d7f-b9af-f76c4e096857","keyframes":{"transform":["scale(0.15)","scale(1)"]},"delay":0,"duration":600,"easing":"cubic-bezier(0.4, 0.4, 0.0, 1)","fill":"forwards"}] [{"selector":"#anim-9de193a9-ab04-429a-9fb5-adea0d7f4a57 [data-leaf-element=\"true\"]","keyframes":{"transform":["translate3d(28.35693346242071%, 0, 0)","translate3d(0%, 0, 0)"]},"delay":0,"duration":2000,"easing":"cubic-bezier(.3,0,.55,1)","fill":"both"}]

It seems that the first chips will be sent to Chinese AI clients around the middle or end of Q2 2024. For more details Govindhtech.com

[{"selector":"#anim-aa4f6908-6099-41e3-9c40-cc0079c127f1","keyframes":{"transform":["translate3d(-116.96114%, 0px, 0)","translate3d(0px, 0px, 0)"]},"delay":0,"duration":1000,"easing":"cubic-bezier(.2, 0, .8, 1)","fill":"both"}] [{"selector":"#anim-5da6081b-f199-40cf-bc64-2293b91b3f57","keyframes":{"transform":["rotateZ(-180deg)","rotateZ(0deg)"]},"delay":0,"duration":1000,"easing":"cubic-bezier(.2, 0, .5, 1)","fill":"forwards"}] [{"selector":"#anim-8fe6a3db-826b-4dd3-9f55-5a90048ec34a [data-leaf-element=\"true\"]","keyframes":{"transform":["translate3d(31.35190206075711%, 0, 0)","translate3d(0%, 0, 0)"]},"delay":0,"duration":2000,"easing":"cubic-bezier(.3,0,.55,1)","fill":"both"}]