# DeepInfra > The price floor for open-model inference, with 150+ models and no minimums. Canonical: https://www.subconscious.dev/providers/deepinfra · By The Subconscious Team · Updated September 30, 2026 - Founded: 2022 - Example models: DeepSeek V4 Flash, Llama 3.1 8B - Website: https://deepinfra.com ## Overview DeepInfra is the price floor for open-model inference. Developers treat it as the reference point for what a token should cost, with small models like Llama 3.1 8B at $0.02 per million and DeepSeek V4 Flash at $0.14 in and $0.28 out. The catalog covers 150+ open models across text, image and speech behind a fully OpenAI-compatible API. There are no minimums, setup fees or contracts on the shared API. Part of that price edge comes from quantization. DeepInfra serves DeepSeek V4 Pro in FP4, which it says keeps inference steady under concurrent load, but that deployment caps context at 66K tokens where Fireworks and Novita offer the full 1M at the same blended price. OpenRouter tags some DeepInfra endpoints as FP4 while rivals run FP8, and some reviewers report weaker output unless they pin FP8 variants. DeepInfra raised a $107M Series B in May 2026 backed by NVIDIA and Samsung. ## Upsides - Consistently at or near the lowest per-token price on popular open models. - Broad catalog with fast intake of new Hugging Face releases. ## Core use cases - High-volume, cost-first workloads like bulk extraction, tagging and synthetic data. - Budget backends for consumer chat and roleplay apps. ## Downsides - Heavy default quantization can cut quality and context length, so teams need to check precision per model. - No managed fine-tuning, so training has to happen elsewhere. ## At a glance | Attribute | Value | |---|---| | Model access | Open weights | | Flagship models | DeepSeek V4 Flash, Llama 3.1 8B | | Speed | ~33 tok/s on DeepSeek V4 Pro (FP4) | | Price | From $0.02 per 1M | | Customization | No managed fine-tuning | | Deployment | Shared API, no contracts | | Long context | 66K on FP4 DeepSeek V4 Pro | ## FAQ ### What is DeepInfra? DeepInfra is the price floor for open-model inference. Developers treat it as the reference point for what a token should cost, with small models like Llama 3.1 8B at $0.02 per million and DeepSeek V4 Flash at $0.14 in and $0.28 out. The catalog covers 150+ open models across text, image and speech behind a fully OpenAI-compatible API. There are no minimums, setup fees or contracts on the shared API. ### What is DeepInfra best for? High-volume, cost-first workloads like bulk extraction, tagging and synthetic data; Budget backends for consumer chat and roleplay apps. ### How much does DeepInfra cost? DeepInfra pricing at a glance: From $0.02 per 1M. Rates change often, so check DeepInfra's pricing page before committing. ### How much context does DeepInfra support? DeepInfra's long-context support: 66K on FP4 DeepSeek V4 Pro. ### What are the downsides of DeepInfra? Heavy default quantization can cut quality and context length, so teams need to check precision per model; No managed fine-tuning, so training has to happen elsewhere. ### What are the best alternatives to DeepInfra? Common alternatives include Subconscious, OpenAI, Anthropic, Google Vertex AI, Amazon Bedrock. Each has a head-to-head comparison with DeepInfra on this site. ## Comparisons - [Subconscious vs DeepInfra](https://www.subconscious.dev/compare/subconscious-vs-deepinfra.md) - [OpenAI vs DeepInfra](https://www.subconscious.dev/compare/openai-vs-deepinfra.md) - [Anthropic vs DeepInfra](https://www.subconscious.dev/compare/anthropic-vs-deepinfra.md) - [Google Vertex AI vs DeepInfra](https://www.subconscious.dev/compare/google-vertex-vs-deepinfra.md) - [Amazon Bedrock vs DeepInfra](https://www.subconscious.dev/compare/aws-bedrock-vs-deepinfra.md) - [Together AI vs DeepInfra](https://www.subconscious.dev/compare/together-ai-vs-deepinfra.md) - [Fireworks AI vs DeepInfra](https://www.subconscious.dev/compare/fireworks-vs-deepinfra.md) - [Baseten vs DeepInfra](https://www.subconscious.dev/compare/baseten-vs-deepinfra.md) - [Groq vs DeepInfra](https://www.subconscious.dev/compare/groq-vs-deepinfra.md) - [Cerebras vs DeepInfra](https://www.subconscious.dev/compare/cerebras-vs-deepinfra.md) - [DeepInfra vs Modal](https://www.subconscious.dev/compare/deepinfra-vs-modal.md) - [DeepInfra vs xAI](https://www.subconscious.dev/compare/deepinfra-vs-xai.md) - [DeepInfra vs DeepSeek](https://www.subconscious.dev/compare/deepinfra-vs-deepseek.md) - [DeepInfra vs Moonshot AI](https://www.subconscious.dev/compare/deepinfra-vs-moonshot-ai.md) - [DeepInfra vs Z.ai](https://www.subconscious.dev/compare/deepinfra-vs-z-ai.md) - [DeepInfra vs Alibaba Cloud](https://www.subconscious.dev/compare/deepinfra-vs-alibaba-cloud.md) - [DeepInfra vs Meta](https://www.subconscious.dev/compare/deepinfra-vs-meta.md) - [DeepInfra vs SambaNova](https://www.subconscious.dev/compare/deepinfra-vs-sambanova.md) - [DeepInfra vs Nebius](https://www.subconscious.dev/compare/deepinfra-vs-nebius.md) - [DeepInfra vs fal](https://www.subconscious.dev/compare/deepinfra-vs-fal.md) - [DeepInfra vs Novita AI](https://www.subconscious.dev/compare/deepinfra-vs-novita-ai.md) - [DeepInfra vs Parasail](https://www.subconscious.dev/compare/deepinfra-vs-parasail.md) - [DeepInfra vs Inference.net](https://www.subconscious.dev/compare/deepinfra-vs-inference-net.md) - [DeepInfra vs GMI Cloud](https://www.subconscious.dev/compare/deepinfra-vs-gmi-cloud.md) - [DeepInfra vs Sail Research](https://www.subconscious.dev/compare/deepinfra-vs-sail-research.md) - [DeepInfra vs Morph](https://www.subconscious.dev/compare/deepinfra-vs-morph.md) - [DeepInfra vs Relace](https://www.subconscious.dev/compare/deepinfra-vs-relace.md) - [DeepInfra vs TypeSafe AI](https://www.subconscious.dev/compare/deepinfra-vs-typesafe-ai.md) - [DeepInfra vs StepFun](https://www.subconscious.dev/compare/deepinfra-vs-stepfun.md) - [DeepInfra vs Runware](https://www.subconscious.dev/compare/deepinfra-vs-runware.md) - [DeepInfra vs StreamLake](https://www.subconscious.dev/compare/deepinfra-vs-streamlake.md) - [DeepInfra vs Wafer](https://www.subconscious.dev/compare/deepinfra-vs-wafer.md) - [DeepInfra vs RunInfra](https://www.subconscious.dev/compare/deepinfra-vs-runinfra.md) - [DeepInfra vs Particle.AI](https://www.subconscious.dev/compare/deepinfra-vs-particle-ai.md) ## Sources - [DeepInfra review, ChatForest](https://chatforest.com/reviews/deepinfra-open-source-inference-cloud/) - [DeepSeek V4 Pro provider benchmarks](https://deepinfra.com/blog/deepseek-v4-pro-max-api-benchmarks-latency-throughput-cost) - [RawSignal on DeepInfra](https://rawsignalai.com/directory/developer-apis/deepinfra) Pricing and model lineups change often; figures are a snapshot.