Skip to content
VECTOR WIREAI INTELLIGENCE
UTC

DSBench-Hard leaderboard — the family board.

2 TRACKED VARIANTS · 1 FIRST-CLASS BOARD

One family, 2 tracked prints. Scoring configs and splits of the base board render as sections of this page, each under its own honest label.

● LIVE DATA6 MODELS ON THIS BOARDLEADER CLAUDE OPUS 4.8 AT 71.7

The board — DSBench-HardTOP 6 OF 6 MEASURED · BEST SCORE PER MODEL, DEFAULT VARIANT

#ModelVendorScoreVLast seen
01Claude Opus 4.8Anthropic71.72026-08-31
02DeepSeek-V4-Flash-Vision-ExpDeepSeek63.62026-08-31
03Kimi K3Moonshot632026-08-13
04DeepSeek-V4-FlashDeepSeek59.62026-08-31
05GLM-5.2Z.ai54.52026-08-13
06DeepSeek-V4-ProDeepSeek31.12026-08-13
◆ IV — INDEPENDENTLY VERIFIED · ◆ AA — AGGREGATOR · ◆ VA — VENDOR · ◇ SA — SOURCE-ATTRIBUTED

Variants of this boardSCORING CONFIGS & SPLITS — NOT CITED STANDALONE, SO THEY LIVE HERE

DSBench-FullStack

VARIANT5 models · best score per model
#ModelVendorScoreVLast seen
01Kimi K3Moonshot73.72026-08-13
02Claude Opus 4.8Anthropic71.62026-08-13
03DeepSeek-V4-FlashDeepSeek68.72026-08-23
04GLM-5.2Z.ai61.82026-08-13
05DeepSeek-V4-ProDeepSeek41.82026-08-13
THE FAMILY BOARD · EVERY PRINT LABELED · EVERY SCORE SOURCE-ATTRIBUTED ON ITS OWN LIVE BOARDAll leaderboardsThe frontier boardModels by vendor