Skip to content
VECTOR WIREAI INTELLIGENCE
UTC

WinoGrande leaderboard — the family board.

3 TRACKED VARIANTS · 1 FIRST-CLASS BOARD

One family, 3 tracked prints. Scoring configs and splits of the base board render as sections of this page, each under its own honest label.

● LIVE DATA40 MODELS ON THIS BOARDLEADER HUNYUAN LARGE AT 88.7

The board — WinoGrandeTOP 40 OF 40 MEASURED · BEST SCORE PER MODEL, DEFAULT VARIANT

#ModelVendorScoreVLast seen
01Hunyuan LargeTencent88.72026-05-31
02GPT-4OpenAI87.52026-08-23
03Llama 3.1 Instruct 405BMeta86.72026-05-31
04DeepSeek-V2DeepSeek86.32026-05-31
05DeepSeek-V3.1-BaseDeepSeek85.92026-06-13
06MiMo V2.5 Pro BaseXiaomi85.62026-06-05
07DeepSeek-V3.2-Exp-BaseDeepSeek85.62026-06-13
08MiMo V2.5 ProXiaomi85.62026-08-23
09Command R+ (Apr '24)Cohere85.42026-08-23
10Llama 3.1 70B InstructMeta85.32026-05-31
11Kimi K2 BaseMoonshot85.32026-06-13
12GLM-4.5-BaseZ.ai85.22026-06-05
13Qwen2 Instruct (72B)Alibaba85.12026-08-23
14DeepSeek-V3DeepSeek84.92026-05-01
15MiMo V2.5 BaseXiaomi84.72026-06-05
16Llama 3.1 Nemotron Instruct 70BNVIDIA84.52026-08-23
17MiMo V2 Flash BaseXiaomi83.82026-06-13
18Gemma 2 27B ITGoogle83.72026-08-23
19Qwen2.5 Instruct 72BAlibaba82.32026-05-01
20Mistral Large 3 675B Base 2512Mistral82.12026-06-05
21DeepSeek-V4-Pro-BaseDeepSeek81.52026-06-05
22Phi 3.5 MoE InstructMicrosoft81.32026-08-23
23Qwen 2.5 Coder 32B InstructAlibaba80.82026-08-23
24Gemma 2 Instruct (9B)Google80.62026-08-23
25Nemotron 3.5 LightningNVIDIA80.02026-08-11
26DeepSeek-V4-Flash-BaseDeepSeek79.52026-06-05
27Nemotron 3 Ultra 550B A55B BaseNVIDIA79.32026-06-05
28Qwen3.5 35B A3BAlibaba79.22026-08-11
29NVIDIA Nemotron 3 Nano 30B A3BNVIDIA79.22026-08-11
30Step3.5 Flash BaseStepFun79.12026-06-05
31Gemma 4 26B A4BGoogle79.12026-08-11
32Gemma (7B)Google792026-08-29
33Nemotron 3 Super 120B A12BNVIDIA78.92026-08-11
34Mistral 7BMistral78.42026-08-29
35Llama 3 Instruct 8BMeta77.42026-08-29
36Phi 3 Mini 4K InstructMicrosoft71.62026-08-29
37Phi 3.5 Mini InstructMicrosoft68.52026-08-23
38Phi 4 Mini InstructMicrosoft672026-08-23
39Gemma 3n E2B InstructGoogle66.82026-09-10
40ERNIE 4.5Baidu51.32026-08-23
◆ IV — INDEPENDENTLY VERIFIED · ◆ AA — AGGREGATOR · ◆ VA — VENDOR · ◇ SA — SOURCE-ATTRIBUTED

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

Other tracked prints

THIN BOARDSunder 5 published models each — listed, never hidden
PrintLeaderScoreModels
WinoGrande (Acc.)DeepSeek-V286.34
WinoGrande (EM)DeepSeek-V4-Pro-Base81.53
THE FAMILY BOARD · EVERY PRINT LABELED · EVERY SCORE SOURCE-ATTRIBUTED ON ITS OWN LIVE BOARDAll leaderboardsThe frontier boardModels by vendor