Model Comparison

DeepSeek-V2.5 vs Claude 3 Sonnet

DeepSeek-V2.5 significantly outperforms across most benchmarks. DeepSeek-V2.5 is 34.3x cheaper per token.

Want to compare interactively?Try the playground

Performance Benchmarks

Comparative analysis across standard metrics

4 benchmarks

DeepSeek-V2.5 outperforms in 4 benchmarks (GSM8k, HumanEval, MATH, MMLU), while Claude 3 Sonnet is better at 0 benchmarks.

DeepSeek-V2.5 significantly outperforms across most benchmarks.

Tue Mar 31 2026 • llm-stats.com

Arena Performance

Human preference votes

Pricing Analysis

Price comparison per million tokens

DeepSeek-V2.5 costs less

For input processing, DeepSeek-V2.5 ($0.14/1M tokens) is 21.4x cheaper than Claude 3 Sonnet ($3.00/1M tokens).

For output processing, DeepSeek-V2.5 ($0.28/1M tokens) is 53.6x cheaper than Claude 3 Sonnet ($15.00/1M tokens).

In conclusion, Claude 3 Sonnet is more expensive than DeepSeek-V2.5.*

* Using a 3:1 ratio of input to output tokens

Lowest available price from all providers
Tue Mar 31 2026 • llm-stats.com
DeepSeek
DeepSeek-V2.5
Input tokens$0.14
Output tokens$0.28
Best providerDeepSeek
Anthropic
Claude 3 Sonnet
Input tokens$3.00
Output tokens$15.00
Best providerAnthropic
Notice missing or incorrect data?Start an Issue→

Context Window

Maximum input and output token capacity

Claude 3 Sonnet accepts 200,000 input tokens compared to DeepSeek-V2.5's 8,192 tokens. Claude 3 Sonnet can generate longer responses up to 200,000 tokens, while DeepSeek-V2.5 is limited to 8,192 tokens.

DeepSeek
DeepSeek-V2.5
Input8,192 tokens
Output8,192 tokens
Anthropic
Claude 3 Sonnet
Input200,000 tokens
Output200,000 tokens
Tue Mar 31 2026 • llm-stats.com

Input Capabilities

Supported data types and modalities

Claude 3 Sonnet supports multimodal inputs, whereas DeepSeek-V2.5 does not.

Claude 3 Sonnet can handle both text and other forms of data like images, making it suitable for multimodal applications.

DeepSeek-V2.5

Text
Images
Audio
Video

Claude 3 Sonnet

Text
Images
Audio
Video

License

Usage and distribution terms

DeepSeek-V2.5 is licensed under deepseek, while Claude 3 Sonnet uses a proprietary license.

License differences may affect how you can use these models in commercial or open-source projects.

DeepSeek-V2.5

deepseek

Open weights

Claude 3 Sonnet

Proprietary

Closed source

Release Timeline

When each model was launched

DeepSeek-V2.5 was released on 2024-05-08, while Claude 3 Sonnet was released on 2024-02-29.

DeepSeek-V2.5 is 2 months newer than Claude 3 Sonnet.

DeepSeek-V2.5

May 8, 2024

1.9 years ago

2mo newer
Claude 3 Sonnet

Feb 29, 2024

2.1 years ago

Knowledge Cutoff

When training data ends

Neither model specifies a knowledge cutoff date.

Unable to compare the recency of their training data.

No cutoff dates available

Provider Availability

DeepSeek-V2.5 is available from DeepSeek, DeepInfra, Hyperbolic. Claude 3 Sonnet is available from Anthropic, Bedrock, Google.

DeepSeek-V2.5

deepseek logo
DeepSeek
Input Price:Input: $0.14/1MOutput Price:Output: $0.28/1M
deepinfra logo
Deepinfra
Input Price:Input: $0.70/1MOutput Price:Output: $1.40/1M
hyperbolic logo
Hyperbolic
Input Price:Input: $2.00/1MOutput Price:Output: $2.00/1M

Claude 3 Sonnet

anthropic logo
Anthropic
Input Price:Input: $3.00/1MOutput Price:Output: $15.00/1M
bedrock logo
AWS Bedrock
Input Price:Input: $3.00/1MOutput Price:Output: $15.00/1M
google logo
Google
Input Price:Input: $3.00/1MOutput Price:Output: $15.00/1M
* Prices shown are per million tokens

Outputs Comparison

Notice missing or incorrect data?Start an Issue discussion→

Key Takeaways

DeepSeek
Less expensive input tokens
Less expensive output tokens
Has open weights
Higher GSM8k score (95.1% vs 92.3%)
Higher HumanEval score (89.0% vs 73.0%)
Higher MATH score (74.7% vs 43.1%)
Higher MMLU score (80.4% vs 79.0%)
Anthropic
Larger context window (200,000 tokens)
Supports multimodal inputs

Detailed Comparison

AI Model Comparison Table
Feature
DeepSeek
DeepSeek-V2.5
Anthropic
Claude 3 Sonnet

FAQ

Common questions about DeepSeek-V2.5 vs Claude 3 Sonnet

DeepSeek-V2.5 significantly outperforms across most benchmarks. DeepSeek-V2.5 is made by DeepSeek and Claude 3 Sonnet is made by Anthropic. The best choice depends on your use case — compare their benchmark scores, pricing, and capabilities above.
DeepSeek-V2.5 scores GSM8k: 95.1%, MT-Bench: 90.2%, HumanEval: 89.0%, BBH: 84.3%, AlignBench: 80.4%. Claude 3 Sonnet scores ARC-C: 93.2%, GSM8k: 92.3%, HellaSwag: 89.0%, MGSM: 83.5%, BIG-Bench Hard: 82.9%.
DeepSeek-V2.5 is 21.4x cheaper for input tokens. DeepSeek-V2.5 costs $0.14/M input and $0.28/M output via deepseek. Claude 3 Sonnet costs $3.00/M input and $15.00/M output via anthropic.
DeepSeek-V2.5 supports 8K tokens and Claude 3 Sonnet supports 200K tokens. A larger context window lets you process longer documents, conversations, or codebases in a single request.
Key differences include context window (8K vs 200K), input pricing ($0.14 vs $3.00/M), multimodal support (no vs yes), licensing (deepseek vs Proprietary). See the full comparison above for benchmark-by-benchmark results.
DeepSeek-V2.5 is developed by DeepSeek and Claude 3 Sonnet is developed by Anthropic.