Key Takeaways
Claude Sonnet 4 wins:
- Larger context window
- Faster response time
- Higher intelligence benchmark
- Better at coding
- Supports vision
- Has reasoning mode
DeepSeek V3.1 wins:
- Cheaper input tokens
- Cheaper output tokens
- Better at math
Price Advantage
DeepSeek V3.1
Benchmark Advantage
Claude Sonnet 4
Context Window
Claude Sonnet 4
Speed
Claude Sonnet 4
Pricing Comparison
Price Comparison
| Metric | Claude Sonnet 4 | DeepSeek V3.1 | Winner |
|---|---|---|---|
| Input (per 1M tokens) | $3.00 | $0.15 | DeepSeek V3.1 |
| Output (per 1M tokens) | $15.00 | $0.75 | DeepSeek V3.1 |
| Cache Read (per 1M) | $0.30 | N/A | Claude Sonnet 4 |
| Cache Write (per 1M) | $3.75 | N/A | Claude Sonnet 4 |
Using a 3:1 input/output ratio, DeepSeek V3.1 is 95% cheaper overall.
Claude Sonnet 4 Providers
Amazon Bedrock $3.00 (Cheapest)
Google $3.00 (Cheapest)
Anthropic $3.00 (Cheapest)
DeepSeek V3.1 Providers
Chutes $0.20 (Cheapest)
DeepInfra $0.21
SiliconFlow $0.27
Novita $0.27
AtlasCloud $0.30
Benchmark Comparison
8
Benchmarks Compared
2
Claude Sonnet 4 Wins
3
DeepSeek V3.1 Wins
Benchmark Scores
| Benchmark | Claude Sonnet 4 | DeepSeek V3.1 | Winner |
|---|---|---|---|
Intelligence Index Overall intelligence score | 33.0 | 28.1 | |
Coding Index Code generation & understanding | 30.6 | 28.4 | |
Math Index Mathematical reasoning | 38.0 | 49.7 | |
MMLU Pro Academic knowledge | 83.7 | 83.3 | - |
GPQA Graduate-level science | 68.3 | 73.5 | |
LiveCodeBench Competitive programming | 44.9 | 57.7 | |
Aider Real-world code editing | 61.3 | - | - |
AIME Competition math | 40.7 | - | - |
DeepSeek V3.1 shows stronger mathematical reasoning abilities.
Cost vs Quality
X-axis:
Y-axis:
Loading chart...
Claude Sonnet 4
Other models
Context & Performance
Context Window
Claude Sonnet 4
1,000,000
tokens
Max output: 64,000 tokens
DeepSeek V3.1
32,768
tokens
Max output: 7,168 tokens
Claude Sonnet 4 has a 97% larger context window.
Speed Performance
| Metric | Claude Sonnet 4 | DeepSeek V3.1 | Winner |
|---|---|---|---|
| Tokens/second | 69.8 tok/s | 0.0 tok/s | |
| Time to First Token | 1.17s | 0.00s |
Claude Sonnet 4 responds Infinity% faster on average.
Capabilities
Feature Comparison
| Feature | Claude Sonnet 4 | DeepSeek V3.1 |
|---|---|---|
| Vision (Image Input) | ||
| Tool/Function Calls | ||
| Reasoning Mode | ||
| Audio Input | ||
| Audio Output | ||
| PDF Input | ||
| Prompt Caching | ||
| Web Search |
License & Release
| Property | Claude Sonnet 4 | DeepSeek V3.1 |
|---|---|---|
| License | Proprietary | Open Source |
| Author | Anthropic | Deepseek |
| Released | May 2025 | Aug 2025 |
Claude Sonnet 4 Modalities
Input
imagetextfile
Output
text
DeepSeek V3.1 Modalities
Input
text
Output
text