DeepSeek V3 0324 (Mar '25): Intelligence, Performance & Price Analysis
Analysis of DeepSeek's DeepSeek V3 0324 (Mar '25) 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.
This model is an updated version of DeepSeek V3 and was launched on 24 March 2025 as DeepSeek-V3-0324. It is architecturally identical to the original December 2024 version of DeepSeek V3.
DeepSeek has launched a newer model, DeepSeek V3.1 (Non-reasoning). We suggest considering this model instead of DeepSeek V3 0324 (Mar '25). See the following pages for a comparison of DeepSeek V3.1 (Non-reasoning) to other models and DeepSeek V3.1 (Non-reasoning) API provider benchmarks.
Comparison Summary
Intelligence:
DeepSeek V3 0324 (Mar '25) is of higher intelligence compared to average, with Artificial Analysis Intelligence Index of 44.
Price:DeepSeek V3 0324 (Mar '25) is cheaper compared to average with a price of $0.48 per 1M Tokens (blended 3:1).
DeepSeek V3 0324 (Mar '25) Input token price: $0.27, Output token price: $1.10 per 1M Tokens.
Speed:DeepSeek V3 0324 (Mar '25) Input token price: $0.27, Output token price: $1.10 per 1M Tokens.
DeepSeek V3 0324 (Mar '25) is slower compared to average, with a output speed of 19.0 tokens per second.
Latency:DeepSeek V3 0324 (Mar '25) has a lower latency compared to average, taking 2.79s to receive the first token (TTFT).
Context Window:DeepSeek V3 0324 (Mar '25) has a smaller context windows than average, with a context window of 130k tokens.
Highlights
Intelligence
Artificial Analysis Intelligence Index; Higher is better
Speed
Output Tokens per Second; Higher is better
Price
USD per 1M Tokens; Lower is better
Parallel Queries:
Prompt Length:
DeepSeek V3 0324 (Mar '25) Model Details
Comparisons to DeepSeek V3 0324 (Mar '25)
- DeepSeek V3 0324 (Mar '25)
GPT-4.1- gpt-oss-20B (high)
- gpt-oss-120B (high)
- GPT-5 (high)
- o3
- GPT-5 (minimal)
Llama 4 Maverick
Gemini 2.5 Pro- Gemini 2.5 Flash (Reasoning)
Claude 4 Sonnet Thinking
Magistral Small- DeepSeek R1 0528
- DeepSeek V3.1 (Non-reasoning)
- DeepSeek V3.1 (Reasoning)
Grok 4
Solar Pro 2 (Reasoning)
MiniMax M1 80k
Llama Nemotron Super 49B v1.5 (Reasoning)
Kimi K2
EXAONE 4.0 32B (Reasoning)
GLM-4.5
Qwen3 235B 2507 (Reasoning)
Further details
R1 1776
Nova Premier
Phi-4
LFM2 1.2B
Granite 3.3 Instruct 8B
Reka Flash 3
DeepHermes 3 - Mistral 24B Preview
Aya Expanse 32B
Jamba 1.7 Large
OpenChat 3.5 (1210)
DBRX Instruct
Llama 3.1 Tulu3 405B
Arctic Instruct
Yi-Large