Codex vs. GitHub Copilot
Comparison between Codex and GitHub Copilot across the Artificial Analysis Coding Agent Index, including benchmark scores, cost, execution time, and token usage.
For details relating to our methodology, see our methodology page.
Explore other comparisons
vs
Comparison
Side-by-side comparison of Codex and GitHub Copilot.
Coding Agent Comparison
Metric |  CodexGPT-6 Astra (Max) |  GitHub Copilot Copilot Fusion (Max) | Analysis |
|---|---|---|---|
Agent Harness | Codex | GitHub Copilot | |
Representative Model | GPT-6 Astra (Max) | Copilot Fusion (Max) | |
Coding Agent Index | 55 | 60 | GitHub Copilot has a higher Coding Agent Index than Codex |
FrontierCode v1.1 | 49% | 51% | GitHub Copilot has a higher FrontierCode v1.1 score than Codex |
Terminal-Bench 4.0 | 56% | 71% | GitHub Copilot has a higher Terminal-Bench 4.0 score than Codex |
SWE-Atlas-QnA-AA | 59% | 57% | Codex has a higher SWE-Atlas-QnA-AA score than GitHub Copilot |
Cost per Task | $7.04 | $5.59 | GitHub Copilot has a lower cost per task than Codex |
Time per Task | 24.8m | 36.1m | Codex has a lower time per task than GitHub Copilot |
Turns per Task | 39.4 | 171 | Codex has a lower turns per task than GitHub Copilot |
Token Usage per Task | 3.2M | 7.3M | Codex has a lower token usage per task than GitHub Copilot |
Cache Hit Rate | 93% | 97% | GitHub Copilot has a higher cache hit rate than Codex |
Model Variants
Evaluated model variants for Codex and GitHub Copilot.
Model Variants
| Codex | GPT-6 Astra (Max) | 55 | 49% | 56% | 59% | $7.04 | 24.8m | 3.2M |
| Codex | GPT-6.1 Sol (Max) | 54 | 49% | 53% | 60% | $1.50 | 23.7m | 3.9M |
| Codex | GPT-6.1 Terra (Xhigh) | 48 | 45% | 46% | 54% | $0.69 | 16.4m | 6.9M |
| Codex | GPT-6.1 Terra (Max) | 48 | 42% | 48% | 54% | $1.24 | 26.1m | 12.3M |
| Codex |  DeepSeek V4.1 Flash (Max) | 47 | 53% | 28% | 61% | $0.39 | 38.1m | 27.2M |
| Codex | GPT-6.1 Terra (High) | 46 | 43% | 42% | 53% | $0.54 | 12.7m | 5.6M |
| Codex | GPT-6.1 Terra (Medium) | 42 | 44% | 35% | 47% | $0.39 | 10.2m | 4.1M |
| Codex | GPT-6.1 Terra (Non-reasoning) | 39 | 46% | 25% | 46% | $0.45 | 9.5m | 5.8M |
| Codex | GPT-6.1 Terra (Low) | 35 | 41% | 25% | 39% | $0.20 | 6.3m | 1.9M |
| Codex | GPT-6 Luna (Max) | 31 | 40% | 16% | 38% | $0.16 | 21.3m | 8.9M |
| GitHub Copilot | Copilot Fusion (Max) | 60 | 51% | 71% | 57% | $5.59 | 36.1m | 7.3M |
Performance
Performance across the Artificial Analysis Coding Agent Index.
Artificial Analysis Coding Agent Index
Artificial Analysis Coding Agent Index v2.0 incorporates 3 benchmarks: FrontierCode v1.1, Terminal-Bench 4.0, and SWE-Atlas-QnA-AA · Higher is better
Color by
Artificial Analysis Coding Agent Index vs. Cost per Task
Artificial Analysis Coding Agent Index vs. average pay-per-token API cost per task (USD)
Color by
Most attractive quadrant
Token Usage
Token consumption across the Artificial Analysis Coding Agent Index.
Token Usage per Task
Average input, cache, and output tokens per task
Prompt cache hit rates can vary significantly by provider routing, which can materially change effective cost.
Artificial Analysis Coding Agent Index vs. Total Tokens
Artificial Analysis Coding Agent Index vs. average total tokens per task
Color by
Most attractive quadrant
Cost
Pay-per-token API cost across the Artificial Analysis Coding Agent Index, based on current per-token pricing.
Cost per Task
Average pay-per-token API cost per task (USD) · Lower is better
Color by
Execution Time
Active agent runtime across the Artificial Analysis Coding Agent Index.
Time per Task
Average agent wall time per task · Lower is better
Color by
Artificial Analysis Coding Agent Index vs. Execution Time
Artificial Analysis Coding Agent Index vs. average agent wall time per task
Color by
Most attractive quadrant