Run the top AI models using a simple API, pay per use. Low cost, scalable and production ready infrastructure.
text-generation
The DeepSeek R1 model has undergone a minor version upgrade, with the current version being DeepSeek-R1-0528.
text-generation
Qwen3 is the latest generation of large language models in Qwen series, offering a comprehensive suite of dense and mixture-of-experts (MoE) models. Built upon extensive training, Qwen3 delivers groundbreaking advancements in reasoning, instruction-following, agent capabilities, and multilingual support
text-generation
Qwen3 is the latest generation of large language models in Qwen series, offering a comprehensive suite of dense and mixture-of-experts (MoE) models. Built upon extensive training, Qwen3 delivers groundbreaking advancements in reasoning, instruction-following, agent capabilities, and multilingual support
text-generation
Qwen3 is the latest generation of large language models in Qwen series, offering a comprehensive suite of dense and mixture-of-experts (MoE) models. Built upon extensive training, Qwen3 delivers groundbreaking advancements in reasoning, instruction-following, agent capabilities, and multilingual support
text-generation
Qwen3 is the latest generation of large language models in Qwen series, offering a comprehensive suite of dense and mixture-of-experts (MoE) models. Built upon extensive training, Qwen3 delivers groundbreaking advancements in reasoning, instruction-following, agent capabilities, and multilingual support.
text-generation
DeepSeek-Prover-V2, an open-source large language model designed for formal theorem proving in Lean 4, with initialization data collected through a recursive theorem proving pipeline powered by DeepSeek-V3. The cold-start training procedure begins by prompting DeepSeek-V3 to decompose complex problems into a series of subgoals. The proofs of resolved subgoals are synthesized into a chain-of-thought process, combined with DeepSeek-V3's step-by-step reasoning, to create an initial cold start for reinforcement learning.
text-generation
The Llama 4 collection of models are natively multimodal AI models that enable text and multimodal experiences. These models leverage a mixture-of-experts architecture to offer industry-leading performance in text and image understanding. Llama 4 Maverick, a 17 billion parameter model with 128 experts
text-generation
The Llama 4 collection of models are natively multimodal AI models that enable text and multimodal experiences. These models leverage a mixture-of-experts architecture to offer industry-leading performance in text and image understanding. Llama 4 Scout, a 17 billion parameter model with 16 experts
text-generation
We introduce DeepSeek-R1, which incorporates cold-start data before RL. DeepSeek-R1 achieves performance comparable to OpenAI-o1 across math, code, and reasoning tasks.
text-generation
We introduce DeepSeek-R1, which incorporates cold-start data before RL. DeepSeek-R1 achieves performance comparable to OpenAI-o1 across math, code, and reasoning tasks.
text-generation
Phi-4-reasoning-plus is a state-of-the-art open-weight reasoning model finetuned from Phi-4 using supervised fine-tuning on a dataset of chain-of-thought traces and reinforcement learning. The supervised fine-tuning dataset includes a blend of synthetic prompts and high-quality filtered data from public domain websites, focused on math, science, and coding skills as well as alignment data for safety and Responsible AI. The goal of this approach was to ensure that small capable models were trained with data focused on high quality and advanced reasoning. Phi-4-reasoning-plus has been trained additionally with Reinforcement Learning, hence, it has higher accuracy but generates on average 50% more tokens, thus having higher latency.
text-generation
Llama Guard 4 is a natively multimodal safety classifier with 12 billion parameters trained jointly on text and multiple images. Llama Guard 4 is a dense architecture pruned from the Llama 4 Scout pre-trained model and fine-tuned for content safety classification. Similar to previous versions, it can be used to classify content in both LLM inputs (prompt classification) and in LLM responses (response classification). It itself acts as an LLM: it generates text in its output that indicates whether a given prompt or response is safe or unsafe, and if unsafe, it also lists the content categories violated.
text-generation
QwQ is the reasoning model of the Qwen series. Compared with conventional instruction-tuned models, QwQ, which is capable of thinking and reasoning, can achieve significantly enhanced performance in downstream tasks, especially hard problems. QwQ-32B is the medium-sized reasoning model, which is capable of achieving competitive performance against state-of-the-art reasoning models, e.g., DeepSeek-R1, o1-mini.
text-generation
DeepSeek-V3-0324, a strong Mixture-of-Experts (MoE) language model with 671B total parameters with 37B activated for each token, an improved iteration over DeepSeek-V3.
text-generation
Gemma 3 introduces multimodality, supporting vision-language input and text outputs. It handles context windows up to 128k tokens, understands over 140 languages, and offers improved math, reasoning, and chat capabilities, including structured outputs and function calling. Gemma 3 27B is Google's latest open source model, successor to Gemma 2
text-generation
Gemma 3 introduces multimodality, supporting vision-language input and text outputs. It handles context windows up to 128k tokens, understands over 140 languages, and offers improved math, reasoning, and chat capabilities, including structured outputs and function calling. Gemma 3-12B is Google's latest open source model, successor to Gemma 2
text-generation
Gemma 3 introduces multimodality, supporting vision-language input and text outputs. It handles context windows up to 128k tokens, understands over 140 languages, and offers improved math, reasoning, and chat capabilities, including structured outputs and function calling. Gemma 3-12B is Google's latest open source model, successor to Gemma 2
text-to-speech
Kokoro is an open-weight TTS model with 82 million parameters. Despite its lightweight architecture, it delivers comparable quality to larger models while being significantly faster and more cost-efficient. With Apache-licensed weights, Kokoro can be deployed anywhere from production environments to personal projects.
text-to-speech
Dia directly generates highly realistic dialogue from a transcript. You can condition the output on audio, enabling emotion and tone control. The model can also produce nonverbal communications like laughter, coughing, clearing throat, etc.
text-to-speech
Orpheus TTS is a state-of-the-art, Llama-based Speech-LLM designed for high-quality, empathetic text-to-speech generation. This model has been finetuned to deliver human-level speech synthesis, achieving exceptional clarity, expressiveness, and real-time streaming performances.
text-to-speech
CSM (Conversational Speech Model) is a speech generation model from Sesame that generates RVQ audio codes from text and audio inputs. The model architecture employs a Llama backbone and a smaller audio decoder that produces Mimi audio codes.
text-generation
Phi-4-multimodal-instruct is a lightweight open multimodal foundation model that leverages the language, vision, and speech research and datasets used for Phi-3.5 and 4.0 models. The model processes text, image, and audio inputs, generating text outputs, and comes with 128K token context length. The model underwent an enhancement process, incorporating both supervised fine-tuning, direct preference optimization and RLHF (Reinforcement Learning from Human Feedback) to support precise instruction adherence and safety measures. The languages that each modal supports are the following: - Text: Arabic, Chinese, Czech, Danish, Dutch, English, Finnish, French, German, Hebrew, Hungarian, Italian, Japanese, Korean, Norwegian, Polish, Portuguese, Russian, Spanish, Swedish, Thai, Turkish, Ukrainian - Vision: English - Audio: English, Chinese, German, French, Italian, Japanese, Spanish, Portuguese
text-generation
DeepSeek-R1-Distill-Llama-70B is a highly efficient language model that leverages knowledge distillation to achieve state-of-the-art performance. This model distills the reasoning patterns of larger models into a smaller, more agile architecture, resulting in exceptional results on benchmarks like AIME 2024, MATH-500, and LiveCodeBench. With 70 billion parameters, DeepSeek-R1-Distill-Llama-70B offers a unique balance of accuracy and efficiency, making it an ideal choice for a wide range of natural language processing tasks.
text-generation
DeepSeek-V3, a strong Mixture-of-Experts (MoE) language model with 671B total parameters with 37B activated for each token. To achieve efficient inference and cost-effective training, DeepSeek-V3 adopts Multi-head Latent Attention (MLA) and DeepSeekMoE architectures, which were thoroughly validated in DeepSeek-V2.
text-generation
Llama 3.3-70B Turbo is a highly optimized version of the Llama 3.3-70B model, utilizing FP8 quantization to deliver significantly faster inference speeds with a minor trade-off in accuracy. The model is designed to be helpful, safe, and flexible, with a focus on responsible deployment and mitigating potential risks such as bias, toxicity, and misinformation. It achieves state-of-the-art performance on various benchmarks, including conversational tasks, language translation, and text generation.
text-generation
Llama 3.3-70B is a multilingual LLM trained on a massive dataset of 15 trillion tokens, fine-tuned for instruction-following and conversational dialogue. The model is designed to be helpful, safe, and flexible, with a focus on responsible deployment and mitigating potential risks such as bias, toxicity, and misinformation. It achieves state-of-the-art performance on various benchmarks, including conversational tasks, language translation, and text generation.
text-generation
Mistral Small 3 is a 24B-parameter language model optimized for low-latency performance across common AI tasks. Released under the Apache 2.0 license, it features both pre-trained and instruction-tuned versions designed for efficient local deployment. The model achieves 81% accuracy on the MMLU benchmark and performs competitively with larger models like Llama 3.3 70B and Qwen 32B, while operating at three times the speed on equivalent hardware.
text-generation
Phi-4 is a model built upon a blend of synthetic datasets, data from filtered public domain websites, and acquired academic books and Q&A datasets. The goal of this approach was to ensure that small capable models were trained with data focused on high quality and advanced reasoning.
automatic-speech-recognition
Whisper is a state-of-the-art model for automatic speech recognition (ASR) and speech translation, proposed in the paper "Robust Speech Recognition via Large-Scale Weak Supervision" by Alec Radford et al. from OpenAI. Trained on >5M hours of labeled data, Whisper demonstrates a strong ability to generalise to many datasets and domains in a zero-shot setting. Whisper large-v3-turbo is a finetuned version of a pruned Whisper large-v3. In other words, it's the exact same model, except that the number of decoding layers have reduced from 32 to 4. As a result, the model is way faster, at the expense of a minor quality degradation.
View all models
Sign up for Deep Infra account using github or Login using github
Choose among hundreds of the most popular ML models
Use a simple rest API to call your model.
Model is deployed in multiple regions
Close to the user
Fast network
Autoscaling
Share resources
Pay per use
Simple pricing
No ML Ops needed
Better cost efficiency
Hassle free ML infrastructure
No ML Ops needed
Better cost efficiency
Hassle free ML infrastructure
Fast scaling infrastructure
Maintain low latency
Scale down when not needed
Run costs
Model | Context | $ per 1M input tokens | $ per 1M output tokens |
---|---|---|---|
mixtral-8x7B-chat | 32k | $0.08 | $0.24 |
wizardLM-2-8x22B | 64k | $0.50 | $0.50 |
Llama-3-8B-Instruct | 8k | $0.03 | $0.06 |
Mistral-7B-v3 | 32k | $0.028 | $0.054 |
MythoMax-L2-13b | 4k | $0.065 | $0.065 |
Llama-3-70B-Instruct | 8k | $0.30 | $0.40 |
Llama-3.1-70B-Instruct | 128k | $0.23 | $0.40 |
Llama-3.1-8B-Instruct | 128k | $0.03 | $0.05 |
Llama-3.1-405B-Instruct | 32k | $0.80 | $0.80 |
Model | Context | $ per 1M input tokens | $ per 1M output tokens |
---|---|---|---|
Loading pricing data... |
You can deploy your own model on our hardware and pay for uptime. You get dedicated SXM-connected GPUs (for multi-GPU setups), automatic scaling to handle load fluctuations and a very competitive price. Read More
GPU | Price |
---|---|
Nvidia A100 GPU | $1.50/GPU-hour |
Nvidia H100 GPU | $2.40/GPU-hour |
Nvidia H200 GPU | $3.00/GPU-hour |
Dedicated A100-80GB, H100-80GB & H200-141GB GPUs for your custom LLM needs
Billed in minute granularity
Invoiced weekly
For dedicated instances and DGX H100 clusters with 3.2Tbps bandwidth, please contact us at dedicated@deepinfra.com
Model | Context | $ per 1M input tokens |
---|---|---|
bge-large-en-v1.5 | 512 | $0.01 |
bge-base-en-v1.5 | 512 | $0.005 |
e5-large-v2 | 512 | $0.01 |
e5-base-v2 | 512 | $0.005 |
gte-large | 512 | $0.01 |
gte-base | 512 | $0.005 |
All models run on H100 or A100 GPUs, optimized for inference performance and low latency.
Our system will automatically scale the model to more hardware based on your needs. We limit each account to 200 concurrent requests. If you want more drop us a line
You have to add a card or pre-pay or you won't be able to use our services. An invoice is always generated at the beginning of the month, and also throughout the month if you hit your tier invoicing threshold. You can also set a spending limit to avoid surprises.
Every user is part of a usage tier. As your usage and your spending goes up, we automatically move you to the next usage tier. Every tier has an invoicing threshold. Once reached an invoice is automatically generated.
Tier | Qualification & Invoicing Threshold | |
---|---|---|
Tier 1 | $20 | |
Tier 2 | $100 paid | $100 |
Tier 3 | $500 paid | $500 |
Tier 4 | $2,000 paid | $2,000 |
Tier 5 | $10,000 paid | $10,000 |