Price Per TokenPrice Per Token
Azure OpenAIvsDeepInfra

Azure OpenAI vs DeepInfra

Compare pricing across 1 shared models. Azure OpenAI offers 87 models, DeepInfra offers 71.

8 Ways to Use Fewer Tokens

Get the free PDF guide — practical tips to cut your token usage and API costs. Subscribe to the Price Per Token newsletter and download instantly.

1
Shared Models
0
Azure OpenAI Cheaper
0
DeepInfra Cheaper
1
Same Price

Price Comparison — Shared Models

Model ↑Azure OpenAI Input DeepInfra Input Azure OpenAI Output DeepInfra Output Cheaper
GPT-OSS-120b $0.150 $0.150 $0.600 $0.600Same

Model Coverage

Azure OpenAIOnly on Azure OpenAI(82)
Azure OpenAI$75.00/M
Babbage$0.500/M
ChatGPT-4o$5.00/M
Codex Mini$0.750/M
curie$2.00/M
GPT-4$30.00/M
GPT-4 Turbo$10.00/M
GPT-4.1$2.00/M
GPT-4.1 Mini$0.200/M
GPT-4.1 Nano$0.100/M
GPT-4o$2.50/M
GPT-4o-mini$0.150/M
GPT-5$0.625/M
GPT-5 Chat$1.25/M
GPT-5 Mini$0.250/M
GPT-5 Nano$0.050/M
GPT-5 Pro$15.00/M
GPT-5.1$1.25/M
GPT-5.1 Chat$0.625/M
GPT-5.2$1.75/M
GPT-5.2 Pro$10.50/M
GPT-5.4$2.50/M
GPT-5.4 Mini$0.750/M
GPT-5.4 Nano$0.200/M
GPT-5.4 Pro$30.00/M
Grok 3$3.00/M
Grok 3 Mini$0.250/M
o1$15.00/M
o1 Mini$0.550/M
o1-pro$150.00/M
o3$2.00/M
o3 Mini$1.10/M
o3 Pro$20.00/M
o4 Mini$0.550/M
R1$1.49/M
text-ada-001$0.200/M
Shared(1)
1
models available on both
Azure OpenAIDeepInfra
83 total71 total
DeepInfraOnly on DeepInfra(70)
DeepSeek V3$0.320/M
Gemma 3 12B$0.040/M
Gemma 3 27B$0.080/M
Gemma 3 4B$0.040/M
GLM 4.6$0.430/M
GLM 4.6V$0.300/M
GLM 4.7$0.400/M
GLM 5$0.800/M
GLM 5.1$1.40/M
GPT-OSS-20b$0.030/M
Kimi K2.5$0.450/M
MiniMax M2.5$0.270/M
Mistral Nemo$0.020/M
MythoMax 13B$0.400/M
Phi 4$0.070/M
Qwen3 14B$0.120/M
Qwen3 32B$0.080/M
Qwen3 Max$1.20/M
Qwen3.5 0.8B$0.010/M
Qwen3.5 9B$0.040/M
Qwen3.5-27B$0.260/M
R1 0528$0.500/M

Full Provider Pricing

Frequently Asked Questions

Azure OpenAI and DeepInfra are evenly matched on pricing across 1 shared models.
Azure OpenAI and DeepInfra share 1 model. Azure OpenAI has 82 exclusive models, while DeepInfra has 70 exclusive models.
Azure OpenAI offers 87 models compared to DeepInfra's 71. However, model count alone doesn't determine the better provider — consider pricing, latency, and which specific models you need.