Codex vs. Gemini CLI

Comparação entre Codex e Gemini CLI no Artificial Analysis Coding Agent Index, incluindo pontuações de benchmarks, custo, tempo de execução e uso de tokens.

Para saber mais sobre nossa metodologia, consulte a página de metodologia.

Explore outras comparações
vs

Destaques

Coding Agent Index

Artificial Analysis Coding Agent Index v1.4 · Higher is better

Tempo por tarefa

Average agent wall time per task · Lower is better

Custo por tarefa

Average API cost per task (USD) · Lower is better

Comparação

Comparação lado a lado de Codex e Gemini CLI.

Comparação de agentes de programação

Métrica
Codex
GPT-5.6 Sol (max)
Gemini CLI
Gemini 3.1 Pro (high)
Análise
Harness do agente
Codex
Gemini CLI
Modelo representativo
GPT-5.6 Sol (max)
Gemini 3.1 Pro (high)
Coding Agent Index
65
33
Codex tem um Coding Agent Index mais alto que Gemini CLI
DeepSWE
69%
14%
Codex tem uma pontuação mais alta em DeepSWE que Gemini CLI
Terminal-Bench v2.1
83%
76%
Codex tem uma pontuação mais alta em Terminal-Bench v2.1 que Gemini CLI
SWE-Atlas-QnA
43%
9%
Codex tem uma pontuação mais alta em SWE-Atlas-QnA que Gemini CLI
Custo por tarefa
$6.42
$1.04
Gemini CLI tem um custo por tarefa mais baixo que Codex
Tempo por tarefa
10.2m
11.1m
Codex tem um tempo por tarefa mais baixo que Gemini CLI
Turnos por tarefa
112.3
31
Gemini CLI tem menos turnos por tarefa que Codex
Uso de tokens por tarefa
13.2M
4.7M
Gemini CLI tem um uso de tokens por tarefa mais baixo que Codex
Taxa de acertos de cache
90%
87%
Codex tem uma taxa de acertos de cache mais alta que Gemini CLI

Variantes de modelo

Variantes de modelo avaliadas para Codex e Gemini CLI.

Variantes de modelo

Codex
GPT-5.6 Sol (max)
65
69%
83%
43%
$6.42
10.2m
13.2M
Codex
GPT-5.6 Sol (high)
64
65%
82%
45%
$3.85
6.2m
8M
Codex
GPT-5.6 Sol (xhigh)
63
67%
80%
43%
$4.80
7.3m
9.9M
Codex
GPT-5.6 Sol (medium)
62
64%
81%
40%
$2.81
5.0m
5.8M
Codex
GPT-5.5 (xhigh)
61
64%
83%
36%
$4.75
10.2m
12.2M
Codex
GPT-5.6 Terra (max)
60
67%
78%
36%
$2.70
8.2m
9.6M
Codex
GPT-5.6 Luna (max)
57
63%
75%
33%
$1.51
8.0m
16M
Codex
GPT-5.6 Terra (xhigh)
56
58%
77%
33%
$1.88
6.7m
6.6M
Codex
GPT-5.5 (medium)
55
57%
79%
31%
$2.65
6.4m
6.9M
Codex
GPT-5.6 Sol (low)
55
53%
78%
34%
$1.66
3.5m
3.2M
Codex
GPT-5.6 Terra (high)
55
60%
72%
31%
$1.56
6.0m
5.6M
Codex
GPT-5.6 Luna (xhigh)
53
57%
71%
31%
$1.21
6.9m
12.8M
Codex
GPT-5.6 Luna (high)
52
53%
73%
29%
$0.92
5.7m
9.6M
Codex
DeepSeek V4 Flash 0731 (max)
50
43%
68%
39%
$0.06
14.5m
20.8M
Codex
GPT-5.6 Terra (medium)
48
46%
70%
29%
$0.88
4.0m
3.2M
Codex
GPT-5.6 Sol (none)
43
35%
60%
34%
$1.40
3.3m
3.4M
Codex
GPT-5.6 Luna (medium)
42
37%
62%
27%
$0.44
3.2m
4.3M
Codex
GPT-5.6 Terra (low)
39
30%
63%
23%
$0.49
2.6m
1.7M
Codex
GPT-5.6 Luna (low)
25
10%
50%
15%
$0.19
1.7m
1.4M
Codex
GPT-5.6 Terra (none)
23
13%
37%
19%
$0.36
1.5m
1.1M
Codex
GPT-5.6 Luna (none)
19
6%
33%
17%
$0.33
2.2m
3.5M
Gemini CLI
Gemini 3.1 Pro (high)
33
14%
76%
9%
$1.04
11.1m
4.7M

Desempenho

Desempenho no Artificial Analysis Coding Agent Index.

Artificial Analysis Coding Agent Index

Artificial Analysis Coding Agent Index v1.4 incorporates 3 benchmarks: DeepSWE, Terminal-Bench v2.1, and SWE-Atlas-QnA · Higher is better
Since benchmarking, we have observed a higher rate of content safety filtering on this endpoint.

The Artificial Analysis Coding Agent Index is a composite score built from DeepSWE, Terminal-Bench v2.1, and SWE-Atlas-QnA.

It is useful for quick comparison, but it should be read alongside the per-eval breakdowns. Two agents with similar index values can still have different strengths across repository tasks, terminal workflows, and rubric-based evaluations.

Uso de tokens

Consumo de tokens no Artificial Analysis Coding Agent Index.

Uso de tokens por tarefa

Média de tokens de entrada, cache e saída por tarefa
Prompt cache hit rates can vary significantly by provider routing, which can materially change effective cost.

Non-cached input tokens sent to the model, including prompts, instructions, tool context, and task context that were not served from prompt cache.

Artificial Analysis Coding Agent Index vs. tokens totais

Artificial Analysis Coding Agent Index vs. média de tokens totais por tarefa
Most attractive quadrant

Each point represents a coding-agent variant. Farther left means lower average total token usage per task, while higher on the chart means higher benchmark performance. Agents toward the upper-left achieve stronger results with fewer tokens.

Custo

Custo da API de pagamento por token no Artificial Analysis Coding Agent Index, com base nos preços atuais por token.

Custo por tarefa

Average pay-per-token API cost per task (USD) · Lower is better

This chart shows the average pay-per-token API cost per task across the Artificial Analysis Coding Agent Index, spanning DeepSWE, Terminal-Bench v2.1, and SWE-Atlas-QnA.

Where applicable, that cost model includes standard input pricing, discounted cached-input pricing, separate cache-write charges, and output pricing rather than treating all prompt tokens as if they were billed at the same uncached input rate.

It is intended to show pay-per-token API cost, not consumer plan pricing or the full operational cost of deploying the system in production. Infrastructure, engineering, and supervision costs are not the focus of this metric.

Artificial Analysis Coding Agent Index vs. custo por tarefa

Artificial Analysis Coding Agent Index vs. média do custo da API de pagamento por token por tarefa (USD)
Most attractive quadrant
Pareto line

Each point represents a coding-agent variant. Farther left means lower average cost per task, while higher on the chart means higher benchmark performance. The most efficient agents sit toward the upper-left: stronger results at lower cost.

Tempo de execução

Tempo de execução ativo do agente no Artificial Analysis Coding Agent Index.

Tempo por tarefa

Average agent wall time per task · Lower is better

This chart uses agent wall time: how long the agent process was actively running on each task.

It does not include environment startup, verifier or judge time, or other harness overhead, so it is a cleaner comparison of how long the agent itself was working.

Artificial Analysis Coding Agent Index vs. tempo de execução

Artificial Analysis Coding Agent Index vs. média do tempo de execução do agente por tarefa
Most attractive quadrant

Each point represents a coding-agent variant. Farther left means shorter average agent runtime per task, while higher on the chart means higher benchmark performance. Agents toward the upper-left deliver stronger results in less active agent time.