Skip to content
VECTOR WIREAI INTELLIGENCE
UTC

LiveBench leaderboard — the family board.

8 TRACKED VARIANTS · 7 FIRST-CLASS BOARDS

One family, 8 tracked prints: livebench_agentic_coding, livebench_coding, livebench_data_analysis, livebench_instruction_following, livebench_math and livebench_reasoning 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 DATA37 MODELS ON THIS BOARDLEADER O3-MINI AT 84.6

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

The board — LiveBenchTOP 37 OF 37 MEASURED · BEST SCORE PER MODEL, DEFAULT VARIANT

#ModelVendorScoreVLast seen
01o3-miniOpenAI84.62026-08-23
02GPT-5.5OpenAI80.72026-08-23
03GPT-5.4OpenAI80.32026-08-23
04Gemini 3.1 ProGoogle79.92026-08-23
05Claude Fable 5Anthropic78.32026-08-23
06Claude Opus 4.8Anthropic77.22026-08-23
07Qwen3 235B A22BAlibaba77.12026-08-23
08Claude Opus 4.7Anthropic76.92026-08-23
09Kimi K2 InstructMoonshot76.42026-08-23
10Claude Opus 4.6Anthropic76.32026-08-23
11Claude Opus 4.5Anthropic76.02026-08-23
12Claude Sonnet 4.6Anthropic75.52026-08-23
13Gemini 3.5 FlashGoogle75.02026-08-23
14Qwen3 32BAlibaba74.92026-08-23
15GPT-5.2OpenAI74.82026-08-23
16Claude Sonnet 4Anthropic74.82026-06-15
17Claude Opus 4Anthropic74.62026-06-15
18Qwen3 30B A3BAlibaba74.32026-08-23
19GPT-5.2 CodexOpenAI74.32026-08-23
20Qwen3.7 MaxAlibaba74.32026-08-23
21DeepSeek-V4-ProDeepSeek73.62026-08-23
22Gemini 3 ProGoogle73.42026-08-23
23QwQ 32B PreviewAlibaba73.12026-08-23
24GPT-5.3 CodexOpenAI72.82026-08-23
25Gemini 3 Flash PreviewGoogle72.42026-08-23
26DeepSeek-V3DeepSeek72.42026-06-15
27Kimi K2.6Moonshot72.22026-08-23
28Kimi K2.7 CodeMoonshot71.92026-08-23
29Qwen3.6 PlusAlibaba70.82026-08-23
30GLM-5.1Z.ai70.22026-08-23
31GPT-5.4 nanoOpenAI70.12026-08-23
32MiniMax M3MiniMax70.02026-08-23
33Kimi K2.5Moonshot69.12026-08-23
34Qwen2.5 Instruct 72BAlibaba52.32026-08-23
35o1OpenAI52.32026-08-23
36Phi 4Microsoft47.62026-08-23
37Qwen2.5 Omni 7BAlibaba29.62026-08-23
◆ IV — INDEPENDENTLY VERIFIED · ◆ AA — AGGREGATOR · ◆ VA — VENDOR · ◇ SA — SOURCE-ATTRIBUTED

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

LiveBench by language

PER-LANGUAGE1 single-language boards, one table
LanguageLeaderScoreModels
livebench-languageClaude Fable 590.760
THE FAMILY BOARD · EVERY PRINT LABELED · EVERY SCORE SOURCE-ATTRIBUTED ON ITS OWN LIVE BOARDAll leaderboardsThe frontier boardModels by vendor