Skip to content
VECTOR WIREAI INTELLIGENCE
UTC

MBPP leaderboard — the family board.

5 TRACKED VARIANTS · 4 FIRST-CLASS BOARDS

One family, 5 tracked prints: MBPP+ (EvalPlus-augmented), MBPP+ and MBPP-Sanitized are cited standalone and keep live boards of their own. Scoring configs and splits of the base board render as sections of this page, each under its own honest label.

● LIVE DATA69 MODELS ON THIS BOARDLEADER SARVAM 30B AT 92.7

First-class boards in this familyCITED STANDALONE · EACH KEEPS ITS OWN LIVE BOARD

#BoardModels scored
01MBPPTHIS PAGE69
03MBPP+10

The board — MBPPTOP 69 OF 69 MEASURED · BEST SCORE PER MODEL, DEFAULT VARIANT

#ModelVendorScoreVLast seen
01Sarvam 30BSarvam AI92.72026-08-23
02Qwen 2.5 Coder 32B InstructAlibaba90.22026-08-23
03Kimi K2 BaseMoonshot892026-06-05
04Qwen2.5 Instruct 72BAlibaba88.22026-08-23
05Granite 4.1 8BIBM87.32026-06-15
06Claude 3.5 HaikuAnthropic85.62026-08-25
07Granite 4.1 30BIBM85.52026-06-15
08GPT-4o miniOpenAI84.82026-08-25
09Llama 3.1 Nemotron Nano 8B V1NVIDIA84.62026-08-23
10Qwen2.5 VL 32B InstructAlibaba842026-08-26
11Granite 4.1 30B BaseIBM83.62026-06-15
12Stable-DiffCoder-8B-BaseByteDance83.62026-06-05
13Qwen2.5 Coder Instruct 7BAlibaba83.52026-06-05
14Yi Coder 9B Chat01.AI822026-06-05
15Nemotron 3 Super 120B A12BNVIDIA81.72026-08-11
16GLM-4.5-BaseZ.ai81.62026-06-05
17Granite 4.1 3B BaseIBM81.52026-06-15
18Qwen3 235B A22BAlibaba81.42026-08-23
19MiMo V2 Flash BaseXiaomi812026-06-05
20Phi 3.5 MoE InstructMicrosoft80.82026-08-23
21Qwen2 Instruct (72B)Alibaba80.22026-08-23
22Seed Coder 8B InstructByteDance79.62026-06-05
23Step3.5 Flash BaseStepFun79.42026-06-05
24Qwen2.5 3B BaseAlibaba79.12026-05-03
25Nemotron 3.5 LightningNVIDIA78.62026-08-11
26Llama 4 MaverickMeta77.62026-08-23
27Qwen3 8BAlibaba772026-06-05
28Qwen2 72BAlibaba76.92026-09-10
29DeepSeek-V3.2-Exp-BaseDeepSeek75.62026-06-05
30DeepSeek-V3DeepSeek75.42026-05-01
31DeepSeek-Coder-6.7B-InstructDeepSeek74.92026-06-05
32DeepSeek-V3.1-BaseDeepSeek74.62026-06-05
33Gemma 3 27BGoogle74.42026-08-25
34DeepSeek-V2DeepSeek73.92026-09-10
35NVIDIA Nemotron 3 Nano 30B A3BNVIDIA73.82026-08-11
36Llama 3.1 Instruct 405BMeta73.42026-05-31
37Qwen2.5 Omni 7BAlibaba73.22026-08-23
38Gemma 3 12BGoogle732026-08-23
39Hunyuan LargeTencent72.62026-05-31
40Granite 4.1 3BIBM71.22026-06-15
41Qwen3.5 35B A3BAlibaba70.82026-08-11
42Llama 3 (70B)Meta70.42026-09-10
43DeepSeek-Coder-6.7B-BaseDeepSeek70.22026-06-05
44Llama 3.1 8B InstructMeta70.12026-06-05
45Phi 3 Mini 4K InstructMicrosoft69.82026-08-29
46Phi 3.5 Mini InstructMicrosoft69.62026-08-23
47Llama 3.1 70B InstructMeta68.62026-05-31
48Gemma 4 26B A4BGoogle68.42026-08-11
49Llama 4 ScoutMeta67.82026-08-23
50Llama 3 Instruct 8BMeta67.72026-08-29
51Qwen2 7B InstructAlibaba67.22026-08-23
52MoonlightMoonshot63.82026-06-12
53Granite 4.1 8B BaseIBM63.82026-06-15
54Gemma 3 4BGoogle63.22026-08-23
55Gemma 2 27B ITGoogle62.62026-08-23
56Qwen2.5 3BAlibaba57.12026-06-05
57Qwen3 4B Base DenseAlibaba55.42026-05-03
58Gemma 2 Instruct (9B)Google52.42026-08-23
59Gemma (7B)Google51.52026-08-29
60Mistral 7BMistral51.12026-08-29
61Llama 3.2 3B InstructMeta48.72026-06-05
62Qwen2.5 7B BaseAlibaba48.52026-05-03
63Gemma 3 1B InstructGoogle35.22026-08-23
64Gemma 3 1B Base DenseGoogle24.82026-05-03
65LFM2.5-8B-A1B-DSparkLiquid AI6.92026-08-21
66LFM2.5-8B-A1BLiquid AI6.92026-08-30
67LFM2.5-1.2B-InstructLiquid AI5.52026-08-21
68LFM2.5-2.6BLiquid AI4.72026-08-30
69LFM2.5-2.6B-DSparkLiquid AI4.72026-08-21
◆ 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
MBPP EvalPlus (base)Llama 3.1 Instruct 405B88.64
THE FAMILY BOARD · EVERY PRINT LABELED · EVERY SCORE SOURCE-ATTRIBUTED ON ITS OWN LIVE BOARDAll leaderboardsThe frontier boardModels by vendor