DeepInfra review

DeepInfra ranks third because it is the value leader for teams that care most about the bill. Its pricing page lists per-token rates for DeepSeek, Kimi, GLM, Qwen, Llama and Nemotron models, many with cached input prices and temporary discounts, and small models go very low: Llama 3.1 8B Instruct Turbo is $0.02 per 1M input tokens and $0.04 per 1M output. All LLMs use an OpenAI-compatible chat completions API, so switching means changing the base URL and model name. When you need your own model, DeepInfra rents dedicated A100, H100, H200, B200 and B300 GPUs billed at minute granularity, with an H100 at $2.90 per GPU hour and a B200 at $4.89. The docs add a Batch API, LoRA adapters, sandboxes and guides for LangChain, LlamaIndex and CrewAI. It offers less around reserved throughput and uptime guarantees than the top two, and multi-node clusters go through sales.

Who it is for

Cost-focused developers who want cheap per-token rates and the lowest dedicated H100 price on this list.

Price

Pay per token; Llama 3.1 8B Instruct Turbo at $0.02 per 1M input tokens; dedicated H100 at $2.90 per GPU hour.

$2.90 / GPU-hour
Quoted from the vendor pricing page on November 7, 2026.

Key features

• OpenAI-compatible chat completions API for all LLM models
• Per-token pricing with cached input rates on many models
• Custom LLM deployments on dedicated A100, H100, H200, B200 and B300 GPUs
• Batch API, LoRA adapters and private model deployment
• Integrations listed for the Anthropic SDK, LangChain, LlamaIndex, Vercel AI SDK and CrewAI

Strengths and weaknesses

• Strength: Dedicated H100 at $2.90 per GPU hour, the lowest on this list
• Strength: Frequent discounts on popular models, such as 51% off DeepSeek V4.1 Flash 1017
• Strength: No long-term contracts or upfront costs
• Weakness: No gpt-oss-120B rate on the pricing page
• Weakness: Fewer reserved-capacity and SLA options than Fireworks or Baseten
• Weakness: Dedicated clusters require emailing the sales team

Key facts

• Dedicated H100: $2.90 per GPU hour
• Dedicated B200: $4.89 per GPU hour
• GPU billing: Minute granularity, invoiced weekly
• API base URL: api.deepinfra.com/v1/openai

Where it ranks

DeepInfra is No. 3 in our Best AI Inference Platforms verdict.