Artificial Analysis Coding Agents

Name: Artificial Analysis Coding Agents
Creator: Artificial Analysis

照 Artificial Analysis 发布的原样呈现：它的指标，它的模型。点模型名进本站详情，点箭头回到原榜。

发布方: Artificial Analysis
许可协议: 未声明
更新频率: 持续更新
本站抓取时间: 2026-10-06
条目数: 31 · 2 条未识别

按 Artificial Analysis Coding Agent Index 排名前 12 的模型

Claude Sonnet 5.5 0.684
Claude Opus 5.5 0.660
Gemini 4 Argon 0.638
GPT 6.1 Sol 0.629
Claude Fable 5.1 0.622
Claude Fable 5.1 XHigh + SWE-2 Medium 0.617
GPT 6 Astra 0.616
Claude Opus 5 0.597
GPT-6 Astra XHigh + SWE-2 Medium 0.589
GPT 6 Sol 0.567
Grok 4.7 0.563
GPT 5.6 Sol 0.546

排序依据只看已识别的模型

31 个有分数

#	模型
1	Claude Sonnet 5.5Claude Code · maxAnthropic	0.684	72.0%	66.9%	66.2%	$14.19	5244.97s
2	Claude Opus 5.5Claude Code · maxAnthropic	0.660	68.4%	66.4%	63.1%	$13.04	3867.09s
3	Gemini 4 ArgonAntigravity CLIGoogle	0.638	78.8%	56.5%	56.1%	$5.84	2070.17s
4	GPT 6.1 SolCodex · xhighOpenAI	0.629	73.2%	61.0%	54.5%	$1.04	931.38s
5	Claude Sonnet 5.5Claude Code · xhighAnthropic	0.629	68.4%	62.1%	58.1%	$3.33	1619.01s
6	Claude Fable 5.1Claude Code · unknown · maxAnthropic	0.622	64.3%	64.8%	57.6%	$12.39	2089.90s
7	Claude Fable 5.1 XHigh + SWE-2 MediumDevin Fusion CLI未识别	0.617	63.1%	65.9%	56.1%	$7.90	2146.99s
8	GPT 6 AstraCodex · maxOpenAI	0.616	67.6%	61.8%	55.6%	$7.47	1762.03s
9	GPT 6.1 SolCodex · mediumOpenAI	0.614	72.0%	60.8%	51.5%	$0.70	651.32s
10	GPT 6.1 SolCodex · highOpenAI	0.601	70.5%	59.9%	50.0%	$0.89	799.77s
11	GPT 6.1 SolCodex · maxOpenAI	0.601	69.6%	57.8%	53.0%	$1.55	1463.05s
12	Claude Opus 5Claude Code · maxAnthropic	0.597	62.5%	62.1%	54.5%	$10.79	2515.54s
13	GPT-6 Astra XHigh + SWE-2 MediumDevin Fusion CLI未识别	0.589	67.3%	59.4%	50.0%	$4.54	1481.90s
14	GPT 6.1 SolCodex · lowOpenAI	0.572	67.6%	55.1%	49.0%	$0.50	517.81s
15	GPT 6 SolCodex · maxOpenAI	0.567	69.0%	57.5%	43.4%	$2.99	1338.05s
16	Grok 4.7Grok Build · xhighSpaceXAI	0.563	72.6%	62.9%	33.3%	$8.82	2354.37s
17	Claude Sonnet 5.5Claude Code · highAnthropic	0.550	66.7%	56.5%	41.9%	$1.24	735.18s
18	GPT 5.6 SolCodex · maxOpenAI	0.546	72.3%	54.0%	37.4%	$6.35	1237.70s
19	Muse Spark 1.3Muse Code · maxMeta	0.543	71.7%	59.4%	31.8%	$3.98	1106.85s
20	GLM 5.3OpencodeZ AI	0.536	61.4%	59.4%	39.9%	$4.24	2887.96s
21	Kimi K3Kimi Code CLIKimi	0.519	68.4%	66.1%	21.2%	$5.05	3689.34s
22	Muse Spark 1.3Muse Code 1.0.2 RC · xhighMeta	0.483	73.2%	54.6%	17.2%	$3.47	1098.20s
23	Grok 4.6Grok Build · xhighSpaceXAI	0.470	64.9%	58.3%	17.7%	$3.57	1167.34s
24	Claude Sonnet 5.5Claude Code · mediumAnthropic	0.459	65.5%	44.9%	27.3%	$0.62	510.77s
25	Qwen 3.8 MaxClaude CodeAlibaba	0.433	51.0%	62.1%	16.7%	$3.48	3784.42s
26	GPT 5.6 LunaCodex · maxOpenAI	0.432	66.4%	48.7%	14.6%	$0.44	1412.26s
27	DeepSeek V4 ProCodex · 0813 · maxDeepSeek	0.431	57.2%	61.8%	10.1%	$0.24	2417.55s
28	Claude Sonnet 5.5Claude Code · lowAnthropic	0.421	61.9%	39.0%	25.3%	$0.48	380.86s
29	Gemini 3.8 FlashAntigravity SDK v0.1.12 · highGoogle	0.419	65.8%	45.2%	14.6%	$2.47	702.71s
30	GPT 6 LunaCodex · maxOpenAI	0.411	63.7%	44.4%	15.2%	$0.18	1285.93s
31	DeepSeek V4 FlashCodex · 0731 · maxDeepSeek	0.387	54.3%	51.3%	10.6%	$0.09	1097.04s

每一列测的是什么

Artificial Analysis Coding Agent Index: 越高越好 · 智能体编程
DeepSWE v1.1: 越高越好
SWE-Atlas-QnA: 越高越好
Terminal-Bench v4: 越高越好 · 智能体编程
Cost: 越低越好 · 成本
Execution Time: 越低越好