Loading the page.
AA Index · data Nov 7, 2026vibecoding.tech_
AI agent ranking
AA Coding Agent Index · artificialanalysis.ai · captured Nov 7, 2026 · 24 pairs
Agent + model
Claude CodeOpus 5.5 · max*
61.7
$11.42
61.0 min
Devin Fusion CLIClaude Opus 5.5 max + GPT-6 Luna · high
61.5
$8.00
51.1 min
Antigravity CLIGemini 4 Carbon · high
59.4
$7.74
29.8 min
Claude CodeSonnet 5.5 · max*
59.3
$13.30
79.7 min
Muse CodeMuse Wonder Full · max
58.9
—
34.9 min
Devin Fusion CLIClaude Fable 5.1 xhigh + SWE-2 · medium
58.7
$8.20
37.6 min
Claude CodeFable 5.1 · max*
57.6
$11.53
35.3 min
CodexGPT-6 Astra · max
54.7
$7.04
24.8 min
CodexGPT-6.1 Sol · max
54.1
$1.50
23.7 min
Antigravity CLIGemini 4 Argon · high
53.2
$11.73
34.7 min
Devin Fusion CLIGPT-6 Astra xhigh + SWE-2 · medium
53.1
$4.34
21.2 min
Muse CodeMuse Wonder · max
52
$8.05
24.5 min
OpenCodeGLM-5.3 · max
48.9
$5.23
56.2 min
CodexGPT-6.1 Terra · xhigh
48.2
$0.69
16.4 min
CodexDeepSeek V4.1 Flash · max
47.3
$0.39
38.1 min
Grok BuildGrok 4.7 · xhigh
46.6
$9.48
35.8 min
DeepSeek HarnessDeepSeek V4.1 Flash 1017 · max
46.3
$0.30
21.4 min
Muse CodeMuse Spark 1.3 · max
45.7
$3.91
23.3 min
Claude CodeStep 5 Preview
43.1
$1.84
61.1 min
Kimi Code CLIKimi K3 · max
41.4
$5.17
62.9 min
Claude CodeQwen3.8 · max
39.2
$3.80
76.1 min
Claude CodeHaiku 5.5 · max
37.3
$2.67
70.9 min
CodexGPT-6 Luna · max
31.1
$0.16
21.3 min
Muse CodeMuse Spark Open · xhigh
26.3
$3.54
26.0 min
Claude CodeOpus 4.8 · max
—
—
—
CodexGPT-5.5 · xhigh
—
—
—
Claude CodeOpus 5 · xhigh
—
—
—
Claude CodeFable 5 · max*
—
—
—
Claude CodeGLM-5.3
—
—
—
CodexGPT-5.6 Sol · max
—
—
—
CodexGPT-5.6 Terra · max
—
—
—
Grok BuildGrok 4.6 · high
—
—
—
CodexGPT-5.6 Luna · max
—
—
—
Claude CodeSonnet 5 · max
—
—
—
Grok BuildGrok 4.5 · high
—
—
—
Claude CodeOpus 5 · max
—
—
—
OpenCodeGemini 3.8 Flash · high
—
—
—
Grok BuildGrok 4.6 · xhigh
—
—
—
Claude CodeDeepSeek V4 Pro · high
—
—
—
Claude CodeGLM-5.2
—
—
—
Claude CodeOpus 4.8 · medium
—
—
—
CodexGPT-5.5 · medium
—
—
—
020406080100
aa index + task price/time - artificialanalysis.ai (Nov 7) · t-bench 4.0 - tbench.ai (Nov 7) · deepswe v1.1 (Sep 22) · «—» - setup was not run
How to read the table
the pair and the model are on different tabs
AI agents are measured as an agent+model pair (config in the badge), models on their own; the best config for each, one row.
the benchmark is selected on the right
the bar and rating column follow the selected benchmark; «—» - not run on it, the row moves down. Sort by clicking a column.
price and time - real
averages per task: agents - artificialanalysis measurement, models - the actual deepswe run bill. Not derived from tokens.
