Inference
LiveBench

LiveBench Agentic Coding

Mean score over agentic tasks in live JavaScript, TypeScript and Python repositories, where the model changes a codebase rather than writing one function.

ModelsModels with a result.55
Top resultBest result on this test.77.3%DeepSeek V4.1 Flash Max
Top-3 spreadPoints from first to third.11.2 pts
UpdatedDate the source changed the results.25 Jun 2026

Result and cost

020406080100$0.01$0.03$0.1$0.3$1$3$10$30RESULTUS DOLLARS PER TASK · LOG SCALE

Results

55 results
#ModelResultThe score from the source.CostUS dollars to run one task.
01DeepSeek V4.1 Flash Max77.3%$0.064
02Claude Opus 5.5max effort71.7%$4.48
03Claude Fable 5.1max effort66.1%$6.22
04Claude Opus 5max effort65.2%$5.34
05DeepSeek V4 Flash Vision Exp65.1%$0.109
06Qwen3.8 Max64.6%$1.75
06Smaug Agentic64.6%$1.81
08Muse Spark 1.3xhigh effort64.1%$1.17
09Claude Fable 5max effort62.2%$8.86
10Qwen3.8-Flash-Next61.6%$0.154
11Qwen3.8-27B61.4%$0.947
12Smaug Flash61.1%$0.078
13GLM-5.360.9%$1.6
14Smaug Mini60.8%$1.07
15Claude Sonnet 5xhigh effort59.4%$2.14
16Muse Spark 1.1xhigh effort58.5%$1.73
17Gemini 3.7 Flashhigh effort58.3%$0.833
18Muse Spark 1.2xhigh effort57.6%$3.92
19GPT-6 Astra Max57.3%$2.43
20Grok 4.657.0%$1.79
21GLM-5.3-Flash56.8%$0.111
22Grok 4.556.5%$0.684
23GPT-5.6 Sol Max56.2%$3.4
24GPT-5.6 Terra Max54.9%$1.35
25Union Alpha54.7%$0
26Gemini 3.8 Flashhigh effort54.2%$1.58
27GPT-5.5xhigh effort54.0%$2.18
27Grok 4.7xhigh effort54.0%$4.15
29GPT-5.4xhigh effort53.8%$1.73
30Ox Alpha Max52.6%$0
31GLM-5.251.8%$1.46
32Claude Opus 4.7xhigh effort50.7%$2.64
33Claude Opus 4.8max effort50.5%$4.96
34GPT-5.2high effort50.3%$1.42
35GPT-5.2-Codex49.4%$0.594
35Inklingxhigh effort49.4%$2.23
37Gemini 3.5 Flashhigh effort49.0%$1.02
37Claude Opus 4.6 Thinking Autohigh effort49.0%$1.24
39DeepSeek V4 Pro 042348.8%$0.142
40GPT-5.6 Luna Max48.4%$0.494
41Kimi K2.6 Thinking46.9%$0.717
42GPT-5.4 nanoxhigh effort46.8%$0.529
43Grok Build 0.145.8%$0.258
44Kimi K2.7 Code45.7%$0.395
45Gemini 3.5 Flash-Litehigh effort45.3%$0.33
46Qwen3.7 Max43.6%$0.796
47Gemini 3.6 Flashhigh effort43.4%$1.35
48Claude Sonnet 4.6 Thinking Automedium effort42.6%$0.63
49DeepSeek V4 Flash 042342.2%$0.032
50GPT-5.4 minixhigh effort41.7%$1.18
51Qwen3.6 Plus41.4%$1.55
52MiniMax M340.7%$0.424
53Claude Opus 4.5 Thinking 64Khigh effort39.7%$1.42
54Qwen3.6-27B39.3%$0.813
55Grok 4.318.5%$0.167