diff --git a/packages/db/src/etl/run-overrides.ts b/packages/db/src/etl/run-overrides.ts
index eebd788a2..250a6d79f 100644
--- a/packages/db/src/etl/run-overrides.ts
+++ b/packages/db/src/etl/run-overrides.ts
@@ -93,6 +93,7 @@ export const PURGED_RUNS: ReadonlySet<number> = new Set([
   30405836523, // 2026-07-28 | Reason: No non-DSpark — kimik3-fp4-b300-vllm-agentic AgentX points run without speculative decoding, and Kimi-K3 agentic coding is published DSpark-only (source run of the PR #2397 sweep-reuse ingest)
   32695861783, // 2026-08-24 | Reason: dev image — the dsv4-fp4-b300-sglang-agentic-hicache-mtp AgentX sweep ran on lmsysorg/sglang-staging:dev-cu13-pr-35880, built from the unmerged SGLang draft PR sgl-project/sglang#35880 (source run of the PR #2701 sweep-reuse ingest)
   32863999669, // 2026-08-25 | Reason: incomplete failing run — Run Sweep on main for #2687 ([Power] Require GB multinode telemetry) failed 30 of 125 jobs (GB200/GB300 multi-node dyn-sgl arms), leaving partial data
+  34512984806, // 2026-09-10 | Reason: wrong results — the dsv41flash-fp4-gb300-vllm-agentic-dspark AgentX sweep for PR #2961 launched the vLLM server via srun without --cpus-per-task, and the GB300 Slurm cluster default allocated only 1 CPU core to the entire vLLM server, crippling throughput; infra launch bug on our side, not a vLLM or NVIDIA recipe issue. Fixed by #3017 (--cpus-per-task=144); results to be re-run
 ]);
 
 export const PURGED_RUN_ATTEMPTS: ReadonlyMap<number, ReadonlySet<number>> = new Map([