DeepInfra vs OpenRouter
Same workloads, both price lists, refreshed daily. On shared line items today: DeepInfra is cheaper on 8 of 30 shared models (input or output price), OpenRouter on 8.
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
OpenRouter
One API key for 340+ models across every major provider, with automatic failover and pass-through pricing. The fastest way to hedge provider risk or A/B models without new integrations.
- 340+ models, one integration
- Automatic provider failover
- Pass-through pricing, small platform fee
Shared models, priced by both
List prices refreshed 2026-08-28 ยท input $/1M tokens| Model | DeepInfra in/out | OpenRouter in/out | Cheaper input |
|---|---|---|---|
| Mistral Nemo | $0.019 / $0.030 | $0.019 / $0.030 | tie |
| Llama 3.2 3B | $0.020 / $0.020 | $0.050 / $0.330 | DeepInfra |
| Llama 3.1 8B | $0.030 / $0.050 | $0.050 / $0.080 | DeepInfra |
| GPT-OSS 20B | $0.030 / $0.140 | $0.020 / $0.100 | OpenRouter |
| GPT-OSS 120B | $0.037 / $0.170 | $0.180 / $0.800 | DeepInfra |
| Gemma 3 12B | $0.050 / $0.150 | $0.050 / $0.150 | tie |
| Gemma 3 4B | $0.050 / $0.100 | $0.050 / $0.100 | tie |
| Mistral Small 3 | $0.050 / $0.080 | $0.050 / $0.080 | tie |
| Nemotron 3 Nano | $0.050 / $0.200 | $0.050 / $0.200 | tie |
| GLM-4.7 Flash | $0.060 / $0.400 | $0.070 / $0.400 | DeepInfra |
| Ling-3.0-Flash | $0.060 / $0.180 | $0.021 / $0.063 | OpenRouter |
| Gemma 4 26B A4B | $0.070 / $0.340 | $0.070 / $0.340 | tie |
| Phi-4 | $0.070 / $0.140 | $0.070 / $0.140 | tie |
| Mistral Small 3.2 24B | $0.075 / $0.200 | $0.100 / $0.300 | DeepInfra |
| Qwen3 32B | $0.080 / $0.280 | $0.080 / $0.280 | tie |
| Gemma 3 27B | $0.080 / $0.160 | $0.080 / $0.450 | tie |
| Nemotron 3.5 Lightning | $0.080 / $0.200 | $0.050 / $0.200 | OpenRouter |
| DeepSeek V4 Flash | $0.090 / $0.180 | $0.050 / $0.100 | OpenRouter |
| Qwen3 Next 80B A3B Instruct | $0.090 / $1.10 | $0.100 / $1.10 | DeepInfra |
| Qwen3.6 35B A3B | $0.100 / $0.950 | $0.100 / $0.900 | tie |
| Gemini-2.0-Flash-001 | $0.100 / $0.400 | $0.100 / $0.400 | tie |
| Qwen3.5-9B | $0.100 / $0.150 | $0.100 / $0.150 | tie |
| Qwen3 30B A3B | $0.120 / $0.500 | $0.120 / $0.500 | tie |
| Qwen3 14B | $0.120 / $0.240 | $0.120 / $0.240 | tie |
| Gemma 4 31B | $0.130 / $0.380 | $0.090 / $0.340 | OpenRouter |
| Qwen3 Next 80B A3B Thinking | $0.140 / $1.40 | $0.150 / $1.20 | DeepInfra |
| Qwen3.5 35B A3B | $0.140 / $1.00 | $0.250 / $2.00 | DeepInfra |
| Hy3 | $0.140 / $0.580 | $0.083 / $0.330 | OpenRouter |
| Qwen3 VL 30B A3B Instruct | $0.150 / $0.600 | $0.150 / $0.600 | tie |
| Qwen3 235B A22B | $0.180 / $0.540 | $0.071 / $0.100 | OpenRouter |
Cheaper-on-count is a tally of listed prices, not a quality verdict: throughput, reliability and quantization differ between providers. Disclosure.