Key Takeaways
GPT-4 Turbo wins:
- No clear advantages in compared metrics
GPT-5 Codex wins:
- Cheaper input tokens
- Cheaper output tokens
- Larger context window
- Faster response time
- Higher intelligence benchmark
- Better at coding
- Better at math
Price Advantage
GPT-5 Codex
Benchmark Advantage
GPT-5 Codex
Context Window
GPT-5 Codex
Speed
GPT-5 Codex
Pricing Comparison
Price Comparison
| Metric | GPT-4 Turbo | GPT-5 Codex | Winner |
|---|---|---|---|
| Input (per 1M tokens) | $10.00 | $1.25 | GPT-5 Codex |
| Output (per 1M tokens) | $30.00 | $10.00 | GPT-5 Codex |
| Cache Read (per 1M) | N/A | $125000.00 | GPT-5 Codex |
Using a 3:1 input/output ratio, GPT-5 Codex is 77% cheaper overall.
GPT-4 Turbo Providers
OpenAI $10.00 (Cheapest)
Vercel $10.00 (Cheapest)
Azure $10.00 (Cheapest)
GPT-5 Codex Providers
OpenAI $1.25 (Cheapest)
Benchmark Comparison
8
Benchmarks Compared
0
GPT-4 Turbo Wins
4
GPT-5 Codex Wins
Benchmark Scores
| Benchmark | GPT-4 Turbo | GPT-5 Codex | Winner |
|---|---|---|---|
Intelligence Index Overall intelligence score | 13.7 | 44.5 | |
Coding Index Code generation & understanding | 21.5 | 38.9 | |
Math Index Mathematical reasoning | - | 98.7 | - |
MMLU Pro Academic knowledge | 69.4 | 86.5 | |
GPQA Graduate-level science | - | 83.7 | - |
LiveCodeBench Competitive programming | 29.1 | 84.0 | |
Aider Real-world code editing | 63.9 | - | - |
AIME Competition math | 15.0 | - | - |
GPT-5 Codex significantly outperforms in coding benchmarks.
Cost vs Quality
X-axis:
Y-axis:
Loading chart...
GPT-4 Turbo
Other models
Context & Performance
Context Window
GPT-4 Turbo
128,000
tokens
Max output: 4,096 tokens
GPT-5 Codex
400,000
tokens
Max output: 128,000 tokens
GPT-5 Codex has a 68% larger context window.
Speed Performance
| Metric | GPT-4 Turbo | GPT-5 Codex | Winner |
|---|---|---|---|
| Tokens/second | 27.2 tok/s | 314.7 tok/s | |
| Time to First Token | 0.86s | 10.95s |
GPT-5 Codex responds 1055% faster on average.
Capabilities
Feature Comparison
| Feature | GPT-4 Turbo | GPT-5 Codex |
|---|---|---|
| Vision (Image Input) | ||
| Tool/Function Calls | ||
| Reasoning Mode | ||
| Audio Input | ||
| Audio Output | ||
| PDF Input | ||
| Prompt Caching | ||
| Web Search |
License & Release
| Property | GPT-4 Turbo | GPT-5 Codex |
|---|---|---|
| License | Proprietary | Proprietary |
| Author | OpenAI | OpenAI |
| Released | Apr 2024 | Sep 2025 |
GPT-4 Turbo Modalities
Input
textimage
Output
text
GPT-5 Codex Modalities
Input
textimage
Output
text