SemiAnalysis

@SemiAnalysis_175k followers

Open on X

Top outlier

3.4×

Best organic post

Peak views

128k

Single post

Bookmarks

999

Across 10 organic posts

Last scan

Oct 16

Community cache

Posts 10 posts

#	Date	Post	↓ Outlier	Views	Likes	Reposts	Bookmarks
1	Oct 15	BREAKING NEWS🚨: Google has submitted its FIRST TPUv7 FP8 vLLM results to our Agentic Inference Benchmark and has demonstrated 1.67x better perf per dollar than even GB300 NVL72 FP	3.4×	128k	1.6k	178	440
2	Oct 14	ALERT🚨🚨: @vllm_project, @Inferact & NVIDIA TOOK CO-DESIGN TO THE EXTREME WITH A GB300 NVL72 INFERENCE OPTIMIZATION TO GET 🚀 7.2x 🚀 BETTER PERF PER TCO THAN LITERALLY EVERY OTHE	1.4×	54k	375	44	135
3	Oct 16	SHOCKING: We have confirmed that multiple Neoclouds have been refused access to GTC Berlin due to them adding AMD GPUs, TPUs, and other chips to their cloud offerings Note that thi	1.2×	45k	348	45	94
4	Oct 16	ALERT🚨: Last night, Google just submitted their FIRST TPUv7 Agentic benchmark, and it delivers 1.42x 🚀 better perf per dollar than GB300 NVL72 FP4, even on an external non-Google	1.1×	43k	407	25	82
5	Oct 16	American open weights are back. 🇺🇸 Huge credit to @finkd, @alexandr_wang and the Meta team for putting Muse Spark Open in developers’ hands. Muse Spark Open beats Nemotron 3 Ultr	1.1×	40k	427	37	57
6	Oct 15	China’s frontier models already run on China’s own chips. “Remember, the GLM-5 paper is almost a year old at this point. They say: ‘From the first day, GLM-5 is full-stack adapted	0.9×	35k	116	12	38
7	Oct 16	View post on X	0.9×	32k	215	8	31
8	Oct 15	ASML Is Building a Hybrid Bonding Tool!	0.7×	27k	212	18	75
9	Oct 15	We take chips apart all the way down to the transistor. Andrew Wagner on SemiAnalysis STEEL, our 18,000 square foot teardown lab in Hillsboro, Oregon: what we tear down, how far do	0.5×	20k	54	2	24
10	Oct 15	What is surprising is that Google TorchTPU vLLM on TPUv7 FP8 with just 4 chips was able to beat GB300 NVL72 FP4 on performance per TCO across most of the pareto frontier. At low in	0.4×	15k	81	7	14
11	Oct 16	Sparse attention fixes memory bandwidth, not memory capacity. “The whole idea of HiCache is trying to offload KV cache from the HBM to a lower tier, higher capacity but lower bandw	0.4×	14k	55	4	23
12	Oct 14	Through extreme co-design, GB300 NVL72 uses an innovative copper backplane to expand the scale-up domain to 72 versus 8 on Hopper. This allows optimizations like wide expert parall	0.3×	13k	33	2	6
13	Oct 15	Great work by the Google TPU, TorchTPU, @Inferact, @RedHat, and vLLM teams on their first amazing AgentX result. We are excited to see Google submit many more models soon to demons	0.3×	12k	69	4	13
14	Oct 15	Watch Now:	0.3×	11k	4	1	1
15	Oct 14	Currently, Inferact & NVIDIA have only submitted up to 200 tok/s/user p90 interactivity SLO, but in the coming days, they will also be extending the curve to 150 tok/s/user and 100	0.3×	10k	34	4	9
16	Oct 16	Another comparison is that at medium to high interactivity, TPUv7 FP8 is up to 3.16x better perf per dollar than even GB200 NVL72 FP4, both with MTP enabled, and GB200 is even usin	0.3×	9.6k	25	2	4
17	Oct 16	@finkd @alexandr_wang Link to dashboard (2/2)	0.2×	9.1k	8	0	1
18	Oct 16	Watch Now:	0.2×	8.4k	1	0	1
19	Oct 16	Great work to the Google TPU, TorchTPU, @Inferact, @RedHat, and vLLM teams on their first amazing AgentX result. Through AgentX, TPUv7 will be demonstrating its incredible perf per	0.2×	7.2k	27	1	3
20	Oct 15	Learn more about STEEL and talk to our team:	0.2×	6.8k	3	0	2

Scan another creator

Run this on any account.

XOutlier builds this table for every public profile you open on x.com, with live data and every post, not just the top 50.

Add to Chrome Download .zip