{"rewrite":{"id":"r_dab0b0e13aa464721b359038","clusterId":"c_617e42edd86c7783388eb96a","slug":"nvidia-s-vera-cpu-puts-olympus-core-and-a-10-wide-decode-front-end-on-the-table","model":"deepseek-v4-1-flash","headline":"NVIDIA's Vera CPU Puts Olympus Core And A 10-Wide Decode Front End On The Table","summary":"A Hot Chips session on NVIDIA's Vera server chip walks through the custom Olympus core, which fetches 16 instructions per cycle and decodes and dispatches 10 per cycle across 18 issue ports. The design keeps the decode throughput of the earlier Arm AGI CPU but pairs it with instruction fusion inside the decode queue and different micro-ops, which the writeup reads as an IPC advantage. NVIDIA designed Vera's core itself, where the earlier Grace chip used Arm's Neoverse V2.","whyItMatters":"Vera is NVIDIA's second data center CPU but the first place it puts a fully self-designed core, so the Olympus numbers mark how far the company has moved from the Neoverse V2 it licensed for Grace.","webCardHtml":"\u003cp\u003eNVIDIA\u0026#39;s Hot Chips talk on Vera went past the core count and into the pipeline. Olympus fetches 16 instructions per cycle, decodes and dispatches 10, and issues to 18 ports, with instruction fusion handled inside the decode queue and rename stages that carry move elimination and memory renaming.\u003c/p\u003e\u003cp\u003eDecode throughput matches the earlier Arm AGI CPU, but the micro-ops differ, and the writeup reads Olympus as the stronger core on instructions per cycle. The larger point is structural: Grace used Arm\u0026#39;s Neoverse V2, so Vera is where NVIDIA\u0026#39;s own core work, which ran through Project Denver and the Carmel-based Tegra T194 before continuing on RISC-V control cores, finally reaches a shipping data center part.\u003c/p\u003e","blueskyPost":"NVIDIA built Vera's Olympus core in house, where Grace leaned on Arm's Neoverse V2. The chip now carries its own CPU design through to production instead of adapting Arm's.","twitterPost":"Olympus is NVIDIA's own core. Grace borrowed Arm's Neoverse V2, so Vera is the chip where the CPU design is entirely NVIDIA's.","threadsPost":"NVIDIA designed the Olympus core in Vera itself. Grace used Arm's Neoverse V2. That shift means the decode queue, micro-ops, and instruction fusion are NVIDIA's own calls now, and the IPC claims rest on engineering NVIDIA controls rather than an Arm core it adapted.","newsletterBlurb":"NVIDIA's Hot Chips session on Vera dug into the custom Olympus core: 16 instructions per cycle at fetch, 10 at decode and dispatch, 18 issue ports, and instruction fusion inside the decode queue. Decode throughput matches the earlier Arm AGI CPU, but the micro-ops differ and the writeup reads Olympus as the better core on IPC. Grace relied on Arm's Neoverse V2, so Vera is where NVIDIA's own core design reaches a data center part.","attributionJson":"[{\"source\":\"ASCII.jp\",\"url\":\"https://ascii.jp/elem/000/004/441/4441711/?rss\",\"title\":\"The True Value of NVIDIA's Custom Core \\\"Olympus\\\": Vera CPU Shows an IPC-Focused Heavy Structure and Memory Bandwidth Dependence\"}]","lintFlagsJson":null,"lintHits":0,"costUsd":0,"inputTokens":4676,"outputTokens":703,"status":"published","repairAttempts":0,"nextRepairAt":null,"factsAttemptedAt":1791778934,"createdAt":"2026-10-12T04:18:19.000Z","publishedAt":"2026-10-12T04:21:09.000Z","updatedAt":"2026-10-12T04:21:09.000Z"},"cluster":{"id":"c_617e42edd86c7783388eb96a","canonicalTitle":"NVIDIA独自コア「Olympus」の真価　Vera CPUが見せたIPC全振りの重厚構造とメモリー帯域依存","representativeArticleId":"a_b1286d83e3639a0fc63250db","sourceCount":1,"writtenSourceCount":1,"writeAttempts":0,"isSolo":true,"entitiesJson":"{\"anime_titles\":[],\"manga_titles\":[],\"work_titles\":[],\"studios\":[\"NVIDIA\"],\"people\":[],\"type\":\"news\",\"domain\":\"other\",\"is_roundup\":false}","contentType":"news","status":"published","firstSeenAt":"2026-10-12T03:00:00.000Z","lastSeenAt":"2026-10-12T03:00:00.000Z","updatedAt":"2026-10-12T04:21:06.000Z"},"attribution":[{"source":"ASCII.jp","url":"https://ascii.jp/elem/000/004/441/4441711/?rss","title":"NVIDIA独自コア「Olympus」の真価　Vera CPUが見せたIPC全振りの重厚構造とメモリー帯域依存"}],"entities":{"anime_titles":[],"manga_titles":[],"work_titles":[],"studios":["NVIDIA"],"people":[],"type":"news","domain":"other","is_roundup":false},"keyFacts":null}