DeepSeek V3 vs GPT-4o-mini
API pricing, context window and capabilities for DeepSeek V3 and GPT-4o-mini, side by side. Prices are the cheapest listed serving route per 1M tokens, from LiteLLM and OpenRouter.
Verdict
computed from the published numbersDeepSeek V3 wins on Terminal-Bench, on intelligence and on context window.
Price
Tied
Terminal-Bench
DeepSeek V3
Intelligence
DeepSeek V3
Context window
DeepSeek V3
Key takeaways
who wins on what, from the published numbersDeepSeek V3 wins
- Cheaper output ($0.20 vs $0.60 /1M)
- Larger context window (164K vs 131K)
- Higher max output (164K vs 16K)
- More serving providers (10 vs 6)
GPT-4o-mini wins
- Cheaper input ($0.20 vs $0.15 /1M)
- Supports vision
Cheaper overall
DeepSeek V3
$0.20 + $0.20 /1M
Larger context
DeepSeek V3
164K
More providers
DeepSeek V3
10 providers
Pricing & specs
cheapest listed route · per 1M tokensOn-demand listed rates, OpenRouter routed quotes and Vercel reseller quotes. Enterprise commitments can be materially lower.
| DeepSeek V3 | GPT-4o-mini | |
|---|---|---|
| Input $/1M | $0.20via Hyperbolic | $0.15via OpenAI best |
| Output $/1M | $0.20via Hyperbolic best | $0.60via OpenAI |
| Context window | 164K best | 131K |
| Max output | 164K best | 16K |
| Serving providers | 10 best | 6 |
| Date | Released Dec 26, 2024 | Released Jul 18, 2024 |
Benchmarks
Benchmark scores, side by side| DeepSeek V3 | GPT-4o-mini | |
|---|---|---|
| Intelligence | 8.5 best | 6.7 |
| Terminal-Bench | 17% best | 6% |
| GPQA Diamond | 56% best | 43% |
| Output speed | 43 t/s | 120 t/s best |
| Latency (TTFT) | 1.22s | 0.84s best |
Cost calculator
choose a serving route for each sideInput tokens / month
Output tokens / month
DeepSeek V3
via Hyperbolic
$1.20
GPT-4o-mini
via Gmi
$1.35
Capabilities
feature support, side by side| Feature | DeepSeek V3 | GPT-4o-mini |
|---|---|---|
| Vision | — | ✓ |
| Tool use | ✓ | ✓ |
| Reasoning | — | — |
| Audio | — | — |
| Prompt caching | ✓ | ✓ |
| Structured output | ✓ | ✓ |
| Web search | — | — |
Related comparisons
Frequently asked questions
Which is cheaper, DeepSeek V3 or GPT-4o-mini?
GPT-4o-mini is cheaper for input ($0.20 vs $0.15 per 1M tokens) from OpenAI. DeepSeek V3 is cheaper for output ($0.20 vs $0.60 per 1M tokens) from Hyperbolic.
What is the context window difference between DeepSeek V3 and GPT-4o-mini?
DeepSeek V3 has a 164K-token context window and GPT-4o-mini has 131K tokens, so DeepSeek V3 can take a larger prompt.
How do DeepSeek V3 and GPT-4o-mini compare on benchmarks?
On the tracked benchmarks, DeepSeek V3 leads on Terminal-Bench, and DeepSeek V3 leads on intelligence.
Listed prices from LiteLLM, OpenRouter and Vercel · last checked Oct 14, 2026
Every weekday
AI moves fast. Here's your debrief.
News, analysis, tools, and more.
For people who build with AI