DeepSeek-V2.5: Intelligence, Performance & Price Analysis
Analysis of DeepSeek's DeepSeek-V2.5 and comparison to other AI models across key metrics including quality, price, performance (tokens per second & time to first token), context window & more. Click on any model to compare API providers for that model. For more details including relating to our methodology, see our FAQs.
DeepSeek has launched a newer model, DeepSeek V3 0324 (Mar '25). We suggest considering this model instead of DeepSeek-V2.5. See the following pages for a comparison of DeepSeek V3 0324 (Mar '25) to other models and DeepSeek V3 0324 (Mar '25) API provider benchmarks.
Comparison Summary
Intelligence:
DeepSeek-V2.5 is of lower intelligence compared to average, with Artificial Analysis Intelligence Index of 20.
Price:DeepSeek-V2.5 is cheaper compared to average with a price of $0.17 per 1M Tokens (blended 3:1).
DeepSeek-V2.5 Input token price: $0.14, Output token price: $0.28 per 1M Tokens.
Speed:Latency:Context Window:DeepSeek-V2.5 Input token price: $0.14, Output token price: $0.28 per 1M Tokens.
DeepSeek-V2.5 has a smaller context windows than average, with a context window of 130k tokens.
Highlights
Intelligence
Artificial Analysis Intelligence Index; Higher is better
Estimate (independent evaluation forthcoming)
Speed
Output Tokens per Second; Higher is better
Price
USD per 1M Tokens; Lower is better
Parallel Queries:
Prompt Length:
DeepSeek-V2.5 Model Details
Comparisons to DeepSeek-V2.5
DeepSeek-V2.5
GPT-5 (high)
GPT-4.1
gpt-oss-120B (high)
gpt-oss-20B (high)
o3
GPT-5 (minimal)
Llama 4 Maverick
Gemini 2.5 Pro
Gemini 2.5 Flash (Reasoning)
Claude 4 Sonnet Thinking
Claude 4.1 Opus Thinking
Magistral Small
DeepSeek V3.1 (Reasoning)
DeepSeek V3.1 (Non-reasoning)
DeepSeek R1 0528
Grok Code Fast 1
Grok 4
Solar Pro 2 (Reasoning)
Llama Nemotron Super 49B v1.5 (Reasoning)
Kimi K2
EXAONE 4.0 32B (Reasoning)
GLM-4.5
Qwen3 235B 2507 (Reasoning)
Further details