DeepInfra Inc.
Serverless inference for machine learning models. Requesty routes to 19 DeepInfra Inc. models starting at $0.02 per 1M input tokens with context windows up to 262K tokens. One API key, OpenAI-compatible SDK, no markup.
Flagship model
meta-llama/Llama-3.3-70B-InstructMMLU Pro
68.9%
GPQA
50.5%
HumanEval
88.4%
SWE-Bench
23.3%
All DeepInfra Inc. models
| Model | Context | Max Output | Input/1M | Output/1M | Capabilities | SWE-Bench |
|---|---|---|---|---|---|---|
Kimi K2.5 | 262K | 131K | $0.45 | $2.25 | 👁🔧⚡ | — |
deepseek-ai/DeepSeek-V3.1 | 164K | — | $0.30 | $1.00 | 🔧 | — |
zai-org/GLM-4.5-Air | 131K | 4K | $0.20 | $1.10 | 🔧 | — |
zai-org/GLM-4.5 | 131K | 4K | $0.60 | $2.20 | 🔧 | — |
Qwen/Qwen3-Coder-480B-A35B-Instruct | 262K | — | $0.40 | $1.60 | 🔧 | — |
phi-4 | 16K | — | $0.07 | $0.14 | — | |
Qwen/Qwen2.5-72B-Instruct | 131K | — | $0.23 | $0.40 | 🔧 | — |
Qwen/Qwen3-32B | 40K | — | $0.10 | $0.30 | 🔧 | — |
Qwen/Qwen2.5-Coder-32B-Instruct | 16K | — | $0.07 | $0.16 | 🔧 | — |
Qwen/Qwen3-235B-A22B | 40K | 4K | $0.20 | $0.60 | 🔧 | — |
meta-llama/Llama-3.3-70B-Instruct | 131K | — | $0.23 | $0.40 | 🔧 | 23% |
meta-llama/Meta-Llama-3.1-405B-Instruct | 131K | — | $0.80 | $0.80 | — | |
meta-llama/Llama-3.3-70B-Instruct-Turbo | 131K | — | $0.12 | $0.30 | 🔧 | — |
meta-llama/Llama-3.2-90B-Vision-Instruct | 131K | 4K | $0.35 | $0.40 | — | |
meta-llama/Meta-Llama-3.1-70B-Instruct | 131K | — | $0.23 | $0.40 | 🔧 | — |
deepseek-ai/DeepSeek-V3 | 128K | 8K | $0.85 | $0.90 | 🔧 | 42% |
deepseek-ai/DeepSeek-R1-Distill-Llama-70B | 64K | 8K | $0.23 | $0.69 | — | |
deepseek-ai/DeepSeek-R1 | 64K | 8K | $0.85 | $2.50 | 🔧 | 49% |
meta-llama/Meta-Llama-3.1-8B-Instruct-Turbo | 131K | — | $0.02 | $0.05 | 🔧 | — |
About DeepInfra Inc. on Requesty
How many DeepInfra Inc. models are available through Requesty?
Requesty routes to 19 DeepInfra Inc. models including regional variants, with pricing synced in real time to the upstream provider.
What is the cheapest DeepInfra Inc. model?
The cheapest DeepInfra Inc. model starts at $0.02 per million input tokens. See the pricing column in the table below for full per-model rates.
Does Requesty add markup on DeepInfra Inc. pricing?
No. Requesty passes through exactly what DeepInfra Inc. charges. You pay the same per-token rates as going direct — plus you get smart routing, caching, analytics, and one unified API for 400+ models.
Is my data used to train DeepInfra Inc. models?
DeepInfra Inc.'s terms state that API data is not used for training. See their privacy policy for the authoritative statement.
Where are DeepInfra Inc. models hosted?
DeepInfra Inc. models are hosted in 🇺🇸 US. Some models are available in additional regions through AWS Bedrock, Azure, or Google Vertex AI — filter by region on the DeepInfra Inc. rows in the models explorer.