Atlas

Benchmarks

← All benchmarks

LMArena Text Arena Coding Elo

Code · 2024-04-14

LMArena text-arena coding leaderboard Elo for pairwise human preference votes on coding-related prompts. Higher Elo indicates stronger human-preference performance.

Top models (higher is better)

ModelScore
Claude Fable 51553
Opus 4.71552
Opus 4.61551
Opus 4.81534
Claude Opus 51532
Kimi K31531
Muse Spark 1.11529
GPT-5.6 Sol1528
Sonnet 4.61527
Sonnet 51526
Muse Spark1526
Qwen3.7 Max Preview1525
Opus 4.51523
GPT-5.41521
Gemini 3.6 Flash1521
Gemini 3.1 Pro Preview1521
GPT-5.51520
GPT-5.6 Terra1520
Grok 4.51520
Gemini 3 Pro Preview1519
GLM-5.11518
MiMo-V2.5-Pro1518
Kimi K2.61515
ERNIE 5.11514
Sonnet 4.51513
Qwen3.7-Plus1509
Gemini 3 Flash Preview1508
Gemini 3.5 Flash1508
Gemini 3.5 Flash-Lite1506
GLM-5.21506
Opus 4.11505
MiMo-V2-Pro1503
DeepSeek-V4-Pro1502
GPT-5.6 Luna1501
Kimi K2.51501
Hy31501
Grok 4.11499
Gemma 4 31B1498
MiniMax M31498
GLM-51497
Loading Atlas data…