Price Per TokenPrice Per Token
Anthropic
Anthropic
vs
Deepseek
Deepseek

Claude 3.5 Sonnet vs DeepSeek V3

A detailed comparison of pricing, benchmarks, and capabilities

OpenClaw

Deploy OpenClaw in Under 1 Minute We handle hosting, scaling, and maintenance

114 out of our 303 tracked models have had a price change in February.

Get our weekly newsletter on pricing changes, new releases, and tools.

Key Takeaways

Claude 3.5 Sonnet wins:

  • Larger context window
  • Faster response time
  • Higher intelligence benchmark
  • Better at coding
  • Supports vision

DeepSeek V3 wins:

  • Cheaper input tokens
  • Cheaper output tokens
Price Advantage
DeepSeek V3
Benchmark Advantage
Claude 3.5 Sonnet
Context Window
Claude 3.5 Sonnet
Speed
Claude 3.5 Sonnet

Pricing Comparison

Price Comparison

MetricClaude 3.5 SonnetDeepSeek V3Winner
Input (per 1M tokens)$6.00$0.32 DeepSeek V3
Output (per 1M tokens)$30.00$0.89 DeepSeek V3
Cache Read (per 1M)$0.60N/A Claude 3.5 Sonnet
Cache Write (per 1M)$7.50N/A Claude 3.5 Sonnet
Using a 3:1 input/output ratio, DeepSeek V3 is 96% cheaper overall.

Claude 3.5 Sonnet Providers

Vercel $3.00 (Cheapest)
Amazon Bedrock $6.00

DeepSeek V3 Providers

DeepSeek $0.01 (Cheapest)
DeepInfra $0.32
Novita $0.40
Fireworks $0.56

Benchmark Comparison

7
Benchmarks Compared
1
Claude 3.5 Sonnet Wins
0
DeepSeek V3 Wins

Benchmark Scores

BenchmarkClaude 3.5 SonnetDeepSeek V3Winner
Intelligence Index
Overall intelligence score
15.9--
Coding Index
Code generation & understanding
30.2--
MMLU Pro
Academic knowledge
77.2--
GPQA
Graduate-level science
59.9--
LiveCodeBench
Competitive programming
38.1--
Aider
Real-world code editing
84.248.4
AIME
Competition math
15.7--
Claude 3.5 Sonnet significantly outperforms in coding benchmarks.

Cost vs Quality

X-axis:
Y-axis:
Loading chart...
Claude 3.5 Sonnet
Other models

Context & Performance

Context Window

Claude 3.5 Sonnet
200,000
tokens
Max output: 8,192 tokens
DeepSeek V3
163,840
tokens
Max output: 163,840 tokens
Claude 3.5 Sonnet has a 18% larger context window.

Speed Performance

MetricClaude 3.5 SonnetDeepSeek V3Winner
Tokens/second0.0 tok/sN/A
Time to First Token0.00sN/A

Capabilities

Feature Comparison

FeatureClaude 3.5 SonnetDeepSeek V3
Vision (Image Input)
Tool/Function Calls
Reasoning Mode
Audio Input
Audio Output
PDF Input
Prompt Caching
Web Search

License & Release

PropertyClaude 3.5 SonnetDeepSeek V3
LicenseProprietaryOpen Source
AuthorAnthropicDeepseek
ReleasedOct 2024Dec 2024

Claude 3.5 Sonnet Modalities

Input
textimagefile
Output
text

DeepSeek V3 Modalities

Input
text
Output
text

Related Comparisons

Compare Claude 3.5 Sonnet with:

Compare DeepSeek V3 with:

Frequently Asked Questions

DeepSeek V3 has cheaper input pricing at $0.32/M tokens. DeepSeek V3 has cheaper output pricing at $0.89/M tokens.
Claude 3.5 Sonnet scores higher on coding benchmarks with a score of 30.2, compared to DeepSeek V3's score of N/A.
Claude 3.5 Sonnet has a 200,000 token context window, while DeepSeek V3 has a 163,840 token context window.
Claude 3.5 Sonnet supports vision. DeepSeek V3 does not support vision.