Key Takeaways
DeepSeek V3.1 wins:
- Cheaper input tokens
- Cheaper output tokens
- Larger context window
- Higher intelligence benchmark
- Better at coding
- Better at math
GPT-4 wins:
- Faster response time
Price Advantage
DeepSeek V3.1
Benchmark Advantage
DeepSeek V3.1
Context Window
DeepSeek V3.1
Speed
GPT-4
Pricing Comparison
Price Comparison
| Metric | DeepSeek V3.1 | GPT-4 | Winner |
|---|---|---|---|
| Input (per 1M tokens) | $0.15 | $30.00 | DeepSeek V3.1 |
| Output (per 1M tokens) | $0.75 | $60.00 | DeepSeek V3.1 |
Using a 3:1 input/output ratio, DeepSeek V3.1 is 99% cheaper overall.
DeepSeek V3.1 Providers
Chutes $0.20 (Cheapest)
DeepInfra $0.21
SiliconFlow $0.27
Novita $0.27
AtlasCloud $0.30
GPT-4 Providers
OpenAI $30.00 (Cheapest)
Azure $30.00 (Cheapest)
Benchmark Comparison
7
Benchmarks Compared
2
DeepSeek V3.1 Wins
0
GPT-4 Wins
Benchmark Scores
| Benchmark | DeepSeek V3.1 | GPT-4 | Winner |
|---|---|---|---|
Intelligence Index Overall intelligence score | 28.1 | 12.8 | |
Coding Index Code generation & understanding | 28.4 | 13.1 | |
Math Index Mathematical reasoning | 49.7 | - | - |
MMLU Pro Academic knowledge | 83.3 | - | - |
GPQA Graduate-level science | 73.5 | - | - |
LiveCodeBench Competitive programming | 57.7 | - | - |
Aider Real-world code editing | - | 67.7 | - |
DeepSeek V3.1 significantly outperforms in coding benchmarks.
Cost vs Quality
X-axis:
Y-axis:
Loading chart...
DeepSeek V3.1
Other models
Context & Performance
Context Window
DeepSeek V3.1
32,768
tokens
Max output: 7,168 tokens
GPT-4
8,191
tokens
Max output: 4,096 tokens
DeepSeek V3.1 has a 75% larger context window.
Speed Performance
| Metric | DeepSeek V3.1 | GPT-4 | Winner |
|---|---|---|---|
| Tokens/second | 0.0 tok/s | 25.4 tok/s | |
| Time to First Token | 0.00s | 0.73s |
GPT-4 responds Infinity% faster on average.
Capabilities
Feature Comparison
| Feature | DeepSeek V3.1 | GPT-4 |
|---|---|---|
| Vision (Image Input) | ||
| Tool/Function Calls | ||
| Reasoning Mode | ||
| Audio Input | ||
| Audio Output | ||
| PDF Input | ||
| Prompt Caching | ||
| Web Search |
License & Release
| Property | DeepSeek V3.1 | GPT-4 |
|---|---|---|
| License | Open Source | Proprietary |
| Author | Deepseek | OpenAI |
| Released | Aug 2025 | May 2023 |
DeepSeek V3.1 Modalities
Input
text
Output
text
GPT-4 Modalities
Input
text
Output
text