DeepInfra review
Updated September 29, 2026 · DeployCue Editorial Team
We earn a commission when you sign up through links on this site. Partner status can affect where a brand appears in our rankings.
Key facts
- Input price from
- $0.019 to $10 (Source, checked September 26, 2026)
- Output price from
- $0.030 to $50 (Source, checked September 26, 2026)
- Models offered
- Deepseek-AI/DeepSeek-V4-Pro, Deepseek-AI/DeepSeek-V4.1-Flash, Moonshotai/Kimi-K3, Qwen/Qwen3.8-Max, Meta-llama/Llama-4-Scout-17B-16E-Instruct, Google/gemini-3.1-pro, Google/gemma-4-31B-it, Nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B, Anthropic/claude-opus-5, Microsoft/phi-4, Mistralai/Mistral-Small-3.2-24B-Instruct-2506, Black-forest-labs/FLUX-2-max (Source, checked September 26, 2026)
- Largest context window
- 1,048,576 (Source, checked September 26, 2026)
- Free credits
- 1,000,000,000 free tokens (DeepStart, at DeepSeek V3.1 prices) (Source, checked September 26, 2026)
- OpenAI-compatible API
- Yes (Source, checked September 26, 2026)
- Compliance certifications
- SOC 2, ISO 27001 (Source, checked September 26, 2026)
- Legal entity
- Deep Infra Inc. (Source, checked September 26, 2026)
- Headquarters
- United States (Source, checked September 26, 2026)
- Founded
- 2022 (Source, checked September 26, 2026)
- Support channels
- Email, Phone, Help-centre, Community (Source, checked September 26, 2026)
- Official website
- deepinfra.com
GPU cloud
- GPU models
- NVIDIA A100, NVIDIA H100, NVIDIA H200, NVIDIA B200, NVIDIA B300 (Source, checked September 26, 2026)
- GPU hour from
- $3.69 (Source, checked September 26, 2026)
- Billing granularity
- Per minute (Source, checked September 26, 2026)
- Egress fee
- No egress fees on on-demand GPU instances (Source, checked September 26, 2026)
- Free credits
- 1,000,000,000 free tokens (DeepStart, at DeepSeek V3.1 prices) (Source, checked September 26, 2026)
- Interconnect
- 5th gen NVLink 1.8TB/s, PCIe Gen5 128GB/s (Source, checked September 26, 2026)
- Compliance certifications
- SOC 2, ISO 27001 (Source, checked September 26, 2026)
- Legal entity
- Deep Infra Inc. (Source, checked September 26, 2026)
- Headquarters
- United States (Source, checked September 26, 2026)
- Founded
- 2022 (Source, checked September 26, 2026)
- Support channels
- Email, Phone, Help-centre, Community (Source, checked September 26, 2026)
- Pricing model
- Hourly billing
Serverless
- Price per million requests
- $0.019 to $10 (Source, checked September 26, 2026)
- Compliance certifications
- SOC 2, ISO 27001 (Source, checked September 26, 2026)
- Legal entity
- Deep Infra Inc. (Source, checked September 26, 2026)
- Headquarters
- United States (Source, checked September 26, 2026)
- Founded
- 2022 (Source, checked September 26, 2026)
- Support channels
- Email, Phone, Help-centre, Community (Source, checked September 26, 2026)
Apps and profiles
- Elsewhere
- LinkedInXgithub.com
How DeepInfra compares
DeepInfra vs BytePlus9.4 vs 8.7
DeepInfra vs RunPod9.4 vs 9.5
DeepInfra vs Zhipu AI (Z.ai)9.4 vs 9.4
DeepInfra vs WaveSpeedAI9.4 vs 8.8
DeepInfra vs AIS Cloud9.4 vs 8.1
Frequently asked questions
Which company operates DeepInfra?+
DeepInfra is operated by Deep Infra Inc.. (Source, checked September 26, 2026)
Where is DeepInfra based?+
DeepInfra is based in United States. (Source, checked September 26, 2026)
When was DeepInfra founded?+
DeepInfra was founded in 2022. (Source, checked September 26, 2026)
How can I contact DeepInfra support?+
DeepInfra offers support through Email, Phone, Help-centre, Community. (Source, checked September 26, 2026)
Sources
- deepinfra.com checked September 26, 2026
Backs: Compliance certifications
“DeepInfra is SOC 2 and ISO 27001 certified.”
- deepinfra.com/about checked September 26, 2026
Backs: Founded
“Founded in September 2022”
- deepinfra.com/deepstart checked September 26, 2026
Backs: Free credits
“DeepStart 1,000,000,000 free tokens*”
- deepinfra.com/gpu-instances checked September 26, 2026
Backs: Billing granularity, Egress fee, GPU hour from, Interconnect
“Pay by the minute with no egress fees.”
- deepinfra.com/pricing checked September 26, 2026
Backs: Largest context window, Input price from, Models offered, Output price from, Price per million requests
“DeepSeek-V4-Pro 1024k”
- deepinfra.com/terms checked September 26, 2026
Backs: Headquarters, Legal entity, Support channels
“2625 Middlefield Road #460 Palo Alto, CA 94306 United States”
- docs.deepinfra.com checked September 26, 2026
Backs: GPU models, OpenAI-compatible API
“Run your own fine-tuned LLM on dedicated A100 / H100 / H200 / B200 / B300 with autoscaling.”
- docs.deepinfra.com/gpu-instances/overview.md checked September 26, 2026
Backs: Pricing model
“billed by the hour, only while running”
Compiled from 8 source pages, checked September 29, 2026. We did not open an account.
