DeepInfra vs SambaNova
Same workloads, both price lists, refreshed daily. On shared line items today: DeepInfra is cheaper on 13 of 13 shared models (input or output price), SambaNova on 0.
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
SambaNova
RDU-based inference cloud serving large open models (405B-class) at high speed with per-token pricing..
- Fast large-model serving
- Full-precision open models
Shared models, priced by both
List prices refreshed 2026-08-23 ยท input $/1M tokens| Model | DeepInfra in/out | SambaNova in/out | Cheaper input |
|---|---|---|---|
| Llama 3.2 3B | $0.020 / $0.020 | $0.080 / $0.160 | DeepInfra |
| Llama 3.1 8B | $0.030 / $0.050 | $0.100 / $0.200 | DeepInfra |
| GPT-OSS 120B | $0.050 / $0.450 | $0.220 / $0.590 | DeepInfra |
| Llama-Guard-3-8B | $0.055 / $0.055 | $0.300 / $0.300 | DeepInfra |
| Llama 4 Scout | $0.080 / $0.300 | $0.400 / $0.700 | DeepInfra |
| Qwen3 32B | $0.100 / $0.280 | $0.400 / $0.800 | DeepInfra |
| Llama 4 Maverick | $0.150 / $0.600 | $0.630 / $1.80 | DeepInfra |
| QwQ-32B | $0.150 / $0.400 | $0.500 / $1.00 | DeepInfra |
| DeepSeek R1 Distill Llama 70B | $0.200 / $0.600 | $0.700 / $1.40 | DeepInfra |
| Llama 3.3 70B | $0.230 / $0.400 | $0.600 / $1.20 | DeepInfra |
| DeepSeek-V3.1 | $0.270 / $1.00 | $3.00 / $4.50 | DeepInfra |
| DeepSeek V3 | $0.380 / $0.890 | $3.00 / $4.50 | DeepInfra |
| DeepSeek R1 | $0.700 / $2.40 | $5.00 / $7.00 | DeepInfra |
Cheaper-on-count is a tally of listed prices, not a quality verdict: throughput, reliability and quantization differ between providers. Disclosure.