Compare pricing, benchmarks, and capabilities across 25 AI models
| Model | Provider | Input $/1M↕ | Output $/1M↕ | Context↕ | Intelligence↑ | Speed↕ | Latency | API |
|---|---|---|---|---|---|---|---|---|
DeepSeek R2 ★ | DeepSeek | $0.55 | $2.19 | 128K | 91% | 60 tok/s | — | |
DeepSeek V3.2 Speciale | DeepSeek | — | — | — | 29.4 | — | — | |
DeepSeek V3.2 (Reasoning) | DeepSeek | — | — | — | 41.7 | 29 tok/s | 1.4s | |
DeepSeek V3.2 Exp (Reasoning) | DeepSeek | — | — | — | 32.9 | 30 tok/s | 1.4s | |
DeepSeek R1 0528 (May '25) | DeepSeek | — | — | — | 27.1 | — | — | |
DeepSeek V3.1 Terminus (Reasoning) | DeepSeek | — | — | — | 33.9 | — | — | |
DeepSeek V3.1 (Reasoning) | DeepSeek | — | — | — | 27.7 | — | — | |
DeepSeek V3.2 Exp (Non-reasoning) | DeepSeek | — | — | — | 28.4 | 31 tok/s | 1.3s | |
DeepSeek V3.1 Terminus (Non-reasoning) | DeepSeek | — | — | — | 28.5 | — | — | |
DeepSeek R1 (Jan '25) | DeepSeek | — | — | — | 18.8 | — | — | |
DeepSeek V3.2 (Non-reasoning) | DeepSeek | — | — | — | 32.1 | 30 tok/s | 1.3s | |
DeepSeek V3.1 (Non-reasoning) | DeepSeek | — | — | — | 28.1 | — | — | |
DeepSeek V3 0324 | DeepSeek | — | — | — | 22.3 | — | — | |
DeepSeek R1 Distill Llama 70B | DeepSeek | — | — | — | 16 | 41 tok/s | 0.5s | |
DeepSeek R1 Distill Qwen 14B | DeepSeek | — | — | — | 15.8 | — | — | |
DeepSeek R1 Distill Qwen 32B | DeepSeek | — | — | — | 17.2 | 42 tok/s | 0.5s | |
DeepSeek R1 0528 Qwen3 8B | DeepSeek | — | — | — | 16.4 | — | — | |
DeepSeek R1 Distill Llama 8B | DeepSeek | — | — | — | 12.1 | — | — | |
DeepSeek Coder V2 Lite Instruct | DeepSeek | — | — | — | 8.5 | — | — | |
DeepSeek R1 Distill Qwen 1.5B | DeepSeek | — | — | — | 9.1 | — | — | |
DeepSeek-Coder-V2 | DeepSeek | — | — | — | 10.6 | — | — | |
DeepSeek-V2.5 | DeepSeek | — | — | — | 12.3 | — | — | |
DeepSeek-V2-Chat | DeepSeek | — | — | — | 9.1 | — | — | |
DeepSeek-V2.5 (Dec '24) | DeepSeek | — | — | — | 12.5 | — | — | |
DeepSeek LLM 67B Chat (V1) | DeepSeek | — | — | — | 8.4 | — | — |
Enter your expected usage to compare costs across models
e.g. 1,000,000 = ~750,000 words
Usually 30–50% of input volume
6 models selected
Prices are approximate and may vary. Check provider documentation for current pricing.