DeepInfra vs DeepSeek
Same workloads, both price lists, refreshed daily. On shared line items today: DeepInfra is cheaper on 3 of 5 shared models (input or output price), DeepSeek on 1.
DeepInfra
Consistently the price floor for open-weight models: Llama, Qwen, DeepSeek at rock-bottom per-token rates with an OpenAI-compatible API and per-request GPU billing options..
- Among the cheapest open-model tokens anywhere
- OpenAI-compatible API
- Dedicated GPU deployments
DeepSeek
Frontier-class reasoning and coding models at a fraction of Western lab prices, with off-peak discounts and open weights for self-hosting..
- Extreme price/performance
- Off-peak pricing windows
- Open weights available
Shared models, priced by both
List prices refreshed 2026-08-28 ยท input $/1M tokens| Model | DeepInfra in/out | DeepSeek in/out | Cheaper input |
|---|---|---|---|
| DeepSeek V4 Flash | $0.090 / $0.180 | $0.440 / $1.32 | DeepInfra |
| DeepSeek V3.2 | $0.260 / $0.380 | $0.280 / $0.400 | DeepInfra |
| DeepSeek V3 | $0.320 / $0.890 | $0.270 / $1.10 | DeepSeek |
| DeepSeek R1 | $0.700 / $2.40 | $0.550 / $2.19 | DeepSeek |
| DeepSeek V4 Pro | $1.30 / $2.60 | $1.32 / $3.96 | DeepInfra |
Cheaper-on-count is a tally of listed prices, not a quality verdict: throughput, reliability and quantization differ between providers. Disclosure.