Kimi K2.5 is an open-source, native multimodal agentic model built through continual pretraining on approximately 15 trillion mixed visual and text tokens atop Kimi-K2-Base. It seamlessly integrates vision and language understanding with advanced agentic capabilities, instant and thinking modes, as well as conversational and agentic paradigms.

Kimi-K2.5

GLM-4.7-Flash is a 30B-A3B MoE model. As the strongest model in the 30B class, GLM-4.7-Flash offers a new option for lightweight deployment that balances performance and efficiency.

GLM-4.7-Flash

P-Image is a state-of-the-art real-time generation model  with exceptional text rendering, fine-detail accuracy, and rock-solid prompt adherence. It’s built for instant creativity at high-fidelity images in about one second at a fraction of typical model costs.

p-image

P-Image-Edit is a high-precision image editing model that applies complex transformations, insertions, removals, and style adjustments in under a second. It delivers state-of-the-art accuracy, clean boundaries, and reliable prompt alignment, making multi-step edits fast, consistent, and production-ready.

p-image-Edit

HiggsAudioV2.5 is a high-quality neural text-to-speech (TTS) model designed for natural-sounding voice generation across a wide range of use cases. It focuses on clarity, stable prosody, and consistent pacing, making it suitable for both short prompts and longer narration.

HiggsAudioV2.5

Chatterbox is a family of three state-of-the-art, open-source text-to-speech models by Resemble AI.  We are excited to introduce Chatterbox-Turbo, our most efficient model yet. Built on a streamlined 350M parameter architecture, Turbo delivers high-quality speech with less compute and VRAM than our previous models. We have also distilled the speech-token-to-mel decoder, previously a bottleneck, reducing generation from 10 steps to just one, while retaining high-fidelity audio output.  Paralinguistic tags are now native to the Turbo model, allowing you to use [cough], [laugh], [chuckle], and more to add distinct realism. While Turbo was built primarily for low-latency voice agents, it excels at narration and creative workflows.  If you like the model but need to scale or tune it for higher accuracy, check out our competitively priced TTS service (link).

chatterbox-turbo

NVIDIA Nemotron 3 Nano is an open reasoning model optimized for fast, cost-efficient inference. Built with a hybrid MoE and Mamba architecture and trained on NVIDIA-curated synthetic reasoning data, it delivers strong multi-step reasoning with stable latency and predictable performance for agentic and production workloads.

Nemotron-3-Nano-30B-A3B

DeepSeek-V3.2 is a large language model designed to harmonize high computational efficiency with strong reasoning and agentic tool-use performance. It introduces DeepSeek Sparse Attention (DSA), a fine-grained sparse attention mechanism that reduces training and inference cost while preserving quality in long-context scenarios. A scalable reinforcement learning post-training framework further improves reasoning, with reported performance in the GPT-5 class, and the model has demonstrated gold-medal results on the 2025 IMO and IOI. V3.2 also uses a large-scale agentic task synthesis pipeline to better integrate reasoning into tool-use settings, boosting compliance and generalization in interactive environments.

DeepSeek-V3.2

Developed by Anthropic, Claude is a family of highly performant, trustworthy AI models built for complex reasoning, advanced coding, and nuanced language understanding. The latest Claude 4 generation delivers breakthrough capabilities in analytical thinking, with Claude 4 Opus setting new standards for intelligence and Claude 4 Sonnet providing exceptional performance with remarkable efficiency.

Claude models excel at understanding context, following complex instructions, and maintaining coherent conversations across extended interactions. With advanced features like extended thinking for deeper reasoning, prompt caching that reduces costs by up to 90%, vision capabilities for image analysis, and robust safety measures, Claude is designed for enterprise applications that demand both sophistication and reliability.

Available with comprehensive API features including streaming responses, batch processing for 50% cost savings, multilingual support across dozens of languages, and flexible context windows up to 200K tokens (1M in beta), Claude is perfect for building intelligent applications like customer support agents, content analysis systems, coding assistants, and complex reasoning workflows that require both accuracy and trustworthiness.

Claude AI family: Claude 4 Opus for complex reasoning, Claude 4 Sonnet for balanced performance, plus advanced capabilities like extended thinking, prompt caching, vision analysis, and enterprise-grade safety APIs.

Claude AI APIs via DeepInfra

claude

Developed by Anthropic, Claude is a family of highly performant, trustworthy AI models built for complex reasoning, advanced coding, and nuanced language understanding

DeepInfra provides access to Anthropic's latest Claude models, featuring the most advanced reasoning capabilities and balanced performance options, all with enterprise-grade safety and reliability.

Claude

DeepSeek develops advanced foundation models optimized for computational efficiency and strong generalization across diverse tasks. The architecture incorporates recent advances in transformer-based systems, delivering robust performance in both zero-shot and fine-tuned scenarios. Models are pretrained on rigorously filtered multilingual corpora with specialized optimizations for mathematical reasoning and algorithmic tasks. The inference stack achieves competitive throughput while maintaining low latency, making it suitable for production deployment. Researchers and engineers can leverage these models for tasks ranging from natural language processing to complex analytical problem-solving.

deepseek

DeepSeek's models are a suite of advanced AI systems that prioritize efficiency, scalability, and real-world applicability.

DeepSeek

Developed by Black Forest Labs (the original creators behind Stable Diffusion), Flux is a family of state-of-the-art image generation and editing models that deliver exceptional visual quality with breakthrough prompt accuracy and photorealism. Built on advanced 12 billion parameter architecture, Flux models excel at understanding exactly what you want to create or modify.

The revolutionary FLUX.1 Kontext introduces game-changing image editing capabilities—simply describe what you want to change in an existing image, and it makes precise modifications while keeping everything else intact. Character faces, lighting, and composition remain consistent across multiple edits, enabling truly iterative creative workflows.

Flux offers specialized variants for every need: Pro delivers maximum quality, Dev provides open-weight flexibility for research, Schnell generates images in just 1-4 steps for rapid iteration, plus dedicated editing tools for specific tasks. Available from open-source to commercial licensing, Flux is perfect for developers building creative applications, product visualization tools, and next-generation image editing experiences.


Flux AI image generation family: FLUX.1 Kontext for in-context editing, FLUX.1 Pro/Dev for text-to-image synthesis, plus comprehensive editing tools and state-of-the-art visual generation APIs.

Flux Image Generation APIs via DeepInfra

flux

Developed by Black Forest Labs, Flux is a family of state-of-the-art image generation and editing models that deliver exceptional visual quality with breakthrough prompt accuracy and photorealism.

DeepInfra provides access to Black Forest Labs' complete Flux ecosystem, offering everything from lightning-fast generation to sophisticated in-context editing capabilities with industry-leading prompt adherence and visual quality.

Flux

Developed by Google DeepMind, Gemini is a family of state-of-the-art thinking models with native multimodal capabilities, designed for advanced reasoning, complex problem-solving, and comprehensive understanding across text, audio, video, and images. Built with revolutionary thinking architecture, Gemini models reason through problems step-by-step before responding, delivering enhanced accuracy and performance for sophisticated applications.

Gemini 2.5 Pro sets new standards for complex reasoning and coding excellence, while Gemini 2.5 Flash provides optimal price-performance for high-volume tasks. With massive context windows up to 1 million tokens, native multimodal processing that handles hours of video and audio, and transparent reasoning capabilities that show step-by-step thinking processes, Gemini excels at document analysis, code generation, scientific research, and agentic workflows.

Perfect for building intelligent applications that require deep reasoning, multimodal understanding, long-context processing, and transparent AI decision-making with Google's enterprise-grade reliability and performance.


Gemini AI family: Advanced thinking models with native multimodal processing for text, audio, video, and image understanding APIs

Gemini AI Model APIs via DeepInfra

gemini

Developed by Google DeepMind, Gemini is a family of state-of-the-art thinking models with native multimodal capabilities

DeepInfra provides access to Google's latest Gemini models, featuring advanced thinking capabilities, native multimodal processing, and industry-leading performance for complex reasoning and development tasks.

Gemini

Developed by Meta, Llama (Large Language Model Meta AI) is a family of state-of-the-art open-weight models designed for efficiency and performance. The latest versions feature Mixture-of-Experts (MoE) architectures, enabling cost-effective inference by dynamically activating subsets of parameters. With support for multimodal inputs (text + images) and extended context windows (up to 10M tokens), Llama excels in tasks like code generation, multilingual understanding, and long-form reasoning. The models support FP8 quantization and batch inference, ensuring low-latency, high-throughput performance for production workloads. With permissive licensing and robust tooling (e.g., Llama Guard for safety), Llama is ideal for developers seeking powerful, customizable AI with minimal overhead.

llama

The Llama 4 collection of models are natively multimodal AI models that enable text and multimodal experiences.

Llama 4

Meta Llama 3 are a collection of pretrained and instruction tuned generative text models in 8B, 70B and 405B sizes.

Llama 3

Llama

Developed by Mistral AI, a leading French research lab, Mistral is a family of open-source AI models built for multilingual excellence, advanced reasoning, and cost-effective performance. These models excel at complex reasoning, mathematics, coding, and specialized tasks while offering complete transparency and deployment freedom through open-source licensing.

Mistral Small 3.2 delivers breakthrough efficiency with native fluency in European languages, while specialized variants handle specific needs: Devstral for coding, Voxtral for audio processing, and Mixtral for high-performance tasks. With Apache 2.0 licensing, extensive context windows up to 128K tokens, and comprehensive customization options, Mistral provides enterprise-grade capabilities without vendor lock-in.

Perfect for building multilingual applications, coding assistants, and reasoning systems where you need both powerful performance and complete control over your AI deployment.

Mistral AI model family: Mistral Small 3.2 for efficient performance, Mixtral for specialized tasks, Devstral for coding, plus multilingual reasoning, mathematics, and open-source flexibility APIs.

Mistral AI Model APIs via DeepInfra

mistral

Developed by Mistral AI, a leading French research lab, Mistral is a family of open-source AI models built for multilingual excellence, advanced reasoning, and cost-effective performance

DeepInfra provides access to Mistral AI's comprehensive open-source model ecosystem, from efficient small models to specialized coding and audio processing variants, all with complete Apache 2.0 licensing freedom.

Mistral

Voxtral is a family of audio models with state-of-the-art speech to text capabilities.

Voxtral

The Nemotron family is a group of large language models developed by NVIDIA, specifically engineered to excel at generating high-quality synthetic data for training other, more powerful AI models. Unlike models focused solely on end-user chat or content creation, Nemotron's core strength lies in producing diverse and realistic text-based training examples—including question-answer pairs, instructions, and conversations—that are crucial for the "supervised fine-tuning" stage of AI development. By providing a robust toolkit for creating these datasets, Nemotron acts as a powerful "force multiplier" in the AI training pipeline, enabling developers to build more capable and refined specialized models efficiently and at scale, without relying solely on scarce, human-curated data.

nemotron

NVIDIA Nemotron is a family of open models customized for efficiency, accuracy, and specialized workloads.

Nemotron

Developed by Alibaba Group's Qwen Team, Qwen is a family of state-of-the-art large language and multimodal models designed for comprehensive AI capabilities and multilingual performance. The latest Qwen3 generation features balanced model architectures including reintroduced Mixture-of-Experts (MoE) variants (Qwen3-30B-A3B and Qwen3-235B-A22B) alongside dense models up to 32B parameters, enabling efficient resource utilization through dynamic parameter activation. 

With support for 119 languages and dialects, hybrid thinking modes that seamlessly alternate between reasoning and instruction-following without model switching, and extended context windows (up to 1M tokens in Qwen3-2507), Qwen excels in tasks like multilingual understanding, code generation, agentic workflows, and complex problem-solving. The models utilize advanced Byte-level Byte Pair Encoding with a 151,646-token vocabulary, structured ChatML formatting for conversational interactions, and robust tool calling capabilities with parallel execution support. 

Available in both proprietary and open-weight versions with flexible licensing, comprehensive model variants (Base, Instruct, Thinking, and hybrid modes), and enhanced Model Context Protocol support, Qwen is ideal for developers seeking powerful, multilingual AI systems with sophisticated reasoning capabilities and minimal deployment complexity.

Qwen AI model family: Qwen3 language models, specialized coding & reasoning models, plus state-of-the-art embedding & reranking APIs for search and RAG applications.

Qwen Model APIs via DeepInfra

qwen

Qwen series offers a comprehensive suite of dense and mixture-of-experts models.

DeepInfra provides access to Qwen's latest generation of large language models, offering both specialized coding models and general-purpose AI systems with advanced reasoning capabilities.

Qwen

This model is part of the GLM-V family of models, introduced in the paper GLM-4.1V-Thinking and GLM-4.5V: Towards Versatile Multimodal Reasoning with Scalable Reinforcement Learning.

ChatGPT said:  EmbeddingGemma is a 300M parameter multilingual open embedding model from Google DeepMind, designed for efficient deployment even on low-resource devices, producing high-quality text vector representations for tasks such as search, classification, clustering, and semantic similarity.

This model is a multilingual version of the OpenAI CLIP-ViT-B32 model, which maps text and images to a common dense vector space. It includes a text embedding model that works for 50+ languages and an image encoder from CLIP. The model was trained using Multilingual Knowledge Distillation, where a multilingual DistilBERT model was trained as a student model to align the vector space of the original CLIP image encoder across many languages.

CodeGemma is a collection of lightweight open code models built on top of Gemma. CodeGemma models are text-to-text and text-to-code decoder-only models and are available as a 7 billion pretrained variant that specializes in code completion and code generation tasks, a 7 billion parameter instruction-tuned variant for code chat and instruction following and a 2 billion parameter pretrained variant for fast code completion.

FLUX.1-dev is a state-of-the-art 12 billion parameter rectified flow transformer developed by Black Forest Labs. This model excels in text-to-image generation, providing highly accurate and detailed outputs. It is particularly well-regarded for its ability to follow complex prompts and generate anatomically accurate images, especially with challenging details like hands and faces.

The CLIP model maps text and images to a shared vector space, enabling various applications such as image search, zero-shot image classification, and image clustering. The model can be used easily after installation, and its performance is demonstrated through zero-shot ImageNet validation set accuracy scores. Multilingual versions of the model are also available for 50+ languages.

Qwen3-Coder-480B-A35B-Instruct is the Qwen3's most agentic code model, featuring Significant Performance on Agentic Coding, Agentic Browser-Use and other foundational coding tasks, achieving results comparable to Claude Sonnet.

L3.3-70B-Euryale-v2.3 is a model focused on creative roleplay from Sao10k

Qwen2.5 is a model pretrained on a large-scale dataset of up to 18 trillion tokens, offering significant improvements in knowledge, coding, mathematics, and instruction following compared to its predecessor Qwen2. The model also features enhanced capabilities in generating long texts, understanding structured data, and generating structured outputs, while supporting multilingual capabilities for over 29 languages.

LLaMA2-13B-Tiefighter is a highly creative and versatile language model, fine-tuned for storytelling, adventure, and conversational dialogue. It combines the strengths of multiple models and datasets, including retro-rodeo and choose-your-own-adventure, to generate engaging and imaginative content. With its ability to improvise and adapt to different styles and formats, Tiefighter is perfect for writers, creators, and anyone looking to spark their imagination.

The Meta Llama 3.2 collection of multilingual large language models (LLMs) is a collection of pretrained and instruction-tuned generative models in 1B and 3B sizes (text in/text out).

Automatically identify and segment foreground objects across video frames and generate a mask. No prompts, just a video.

ClarityAI/flux integrates the Flux AI model into the upscaling process, enabling high-resolution enhancements with superior face preservation and support for LoRAs to apply specific styles or identities.

Bria 3.2 is the next-generation commercial-ready text-to-image model. With just 4 billion parameters, it provides exceptional aesthetics and text rendering, evaluated to be on par to leading open-source models, and outperforming other licensed models.

The Wan2.1 1.3B model is a lightweight, efficient text-to-video generator. Despite its compact size, it delivers impressive performance across benchmarks and generates high-quality 480P videos.

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. 

The CLIP model was developed by OpenAI to investigate the robustness of computer vision models. It uses a Vision Transformer architecture and was trained on a large dataset of image-caption pairs. The model shows promise in various computer vision tasks but also has limitations, including difficulties with fine-grained classification and potential biases in certain applications.

Remove unwanted objects or regions from video using a mask, reconstructs the background with intelligent content-aware fill.

Bria GenFill enables high-quality object addition or visual transformation. Trained exclusively on licensed data for safe and risk-free commercial use.

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. 

Llama-3.3-Nemotron-Super-49B-v1.5 is a large language model (LLM) optimized for advanced reasoning, conversational interactions, retrieval-augmented generation (RAG), and tool-calling tasks. Derived from Meta's Llama-3.3-70B-Instruct, it employs a Neural Architecture Search (NAS) approach, significantly enhancing efficiency and reducing memory requirements. 

Multi-reference visual intelligence with unprecedented detail, color precision, and spatial reasoning.  The most advanced image generation and editing model. Generate photorealistic images with precise control.

A drop-in replacement for Flux [Dev] that delivers sharper details, richer colors, and enhanced realism, while instantly boosting LoRAs and LyCORIS with full compatibility.

We present a sentence similarity model based on the Sentence Transformers architecture, which maps sentences to a 384-dimensional dense vector space. The model uses a pre-trained BERT encoder and applies mean pooling on top of the contextualized word embeddings to obtain sentence embeddings. We evaluate the model on the Sentence Embeddings Benchmark.

Light and fast. Remove the background of your videos to bring the foreground elements to focus. No more unwanted distractions.

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

The SDXL Turbo model, developed by Stability AI, is an optimized, fast text-to-image generative model. It is a distilled version of SDXL 1.0, leveraging Adversarial Diffusion Distillation (ADD) to generate high-quality images in less steps.

LLaMa 2 is a collections of LLMs trained by Meta. This is the 70B chat optimized version. This endpoint has per token pricing.

Meet Qwen3-VL — the most powerful vision-language model in the Qwen series to date.  This generation delivers comprehensive upgrades across the board: superior text understanding & generation, deeper visual perception & reasoning, extended context length, enhanced spatial and video dynamics comprehension, and stronger agent interaction capabilities.

Kimi K2 0905 is the September update of Kimi K2 0711. It is a large-scale Mixture-of-Experts (MoE) language model developed by Moonshot AI, featuring 1 trillion total parameters with 32 billion active per forward pass. It supports long-context inference up to 256k tokens, extended from the previous 128k.  This update improves agentic coding with higher accuracy and better generalization across scaffolds, and enhances frontend coding with more aesthetic and functional outputs for web, 3D, and related tasks. Kimi K2 is optimized for agentic capabilities, including advanced tool use, reasoning, and code synthesis. It excels across coding (LiveCodeBench, SWE-bench), reasoning (ZebraLogic, GPQA), and tool-use (Tau2, AceBench) benchmarks. The model is trained with a novel stack incorporating the MuonClip optimizer for stable large-scale MoE training.

Meta developed and released the Meta Llama 3.1 family of large language models (LLMs), a collection of pretrained and instruction tuned generative text models in 8B, 70B and 405B sizes

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.

Bria Enhance improves overall image quality by sharpening details, balancing colors, and boosting clarity for crisp, professional visuals. Trained only on licensed data, it’s safe, reliable, and ready for commercial use.

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.

A sentence similarity model that can be used for various NLP tasks such as text classification, sentiment analysis, named entity recognition, question answering, and more. It utilizes the CoSENT architecture, which consists of a transformer encoder and a pooling module, to encode input texts into vectors that capture their semantic meaning. The model was trained on the nli_zh dataset and achieved high performance on various benchmark datasets.

The Dolphin 2.6 Mixtral 8x7b model is a finetuned version of the Mixtral-8x7b model, trained on a variety of data including coding data, for 3 days on 4 A100 GPUs. It is uncensored and requires trust_remote_code. The model is very obedient and good at coding, but not DPO tuned. The dataset has been filtered for alignment and bias. The model is compliant with user requests and can be used for various purposes such as generating code or engaging in general chat.

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

BGE embedding is a general Embedding Model. It is pre-trained using retromae and trained on large-scale pair data using contrastive learning. Note that the goal of pre-training is to reconstruct the text, and the pre-trained model cannot be used for similarity calculation directly, it needs to be fine-tuned

MiniMax-M2 is a Mini model built for Max coding & agentic workflows with just 10 billion activated parameters

Euryale 70B v2.1 is a model focused on creative roleplay from Sao10k

Most widely used version of Stable Diffusion. Trained on 512x512 images, it can generate realistic images given text description

PaddleOCR-VL is a SOTA and resource-efficient model tailored for document parsing. Its core component is PaddleOCR-VL-0.9B, a compact yet powerful vision-language model (VLM) that integrates a NaViT-style dynamic resolution visual encoder with the ERNIE-4.5-0.3B language model to enable accurate element recognition. This innovative model efficiently supports 109 languages and excels in recognizing complex elements (e.g., text, tables, formulas, and charts), while maintaining minimal resource consumption. Through comprehensive evaluations on widely used public benchmarks and in-house benchmarks, PaddleOCR-VL achieves SOTA performance in both page-level document parsing and element-level recognition. It significantly outperforms existing solutions, exhibits strong competitiveness against top-tier VLMs, and delivers fast inference speeds. These strengths make it highly suitable for practical deployment in real-world scenarios.

Dolphin 2.9.1, a fine-tuned Llama-3-70b model. The new model, trained on filtered data, is more compliant but uncensored. It demonstrates improvements in instruction, conversation, coding, and function calling abilities.

This is a 32B reasoning model trained from Qwen2.5-32B-Instruct with 17K data. The performance is on par with o1-preview model on both math and coding.

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

PixVerse's 720p resolution offers a fast and reliable option for generating standard HD videos, ideal for quick previews and social media content where generation speed is prioritized over maximum detail.

The GTE models are trained by Alibaba DAMO Academy. They are mainly based on the BERT framework and currently offer three different sizes of models, including GTE-large, GTE-base, and GTE-small. The GTE models are trained on a large-scale corpus of relevance text pairs, covering a wide range of domains and scenarios. This enables the GTE models to be applied to various downstream tasks of text embeddings, including information retrieval, semantic textual similarity, text reranking, etc.

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

The Phi-3-Medium-4K-Instruct is a powerful and lightweight language model with 14 billion parameters, trained on high-quality data to excel in instruction following and safety measures. It demonstrates exceptional performance across benchmarks, including common sense, language understanding, and logical reasoning, outperforming models of similar size.

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

The Qwen3 Embedding model series is the latest proprietary model of the Qwen family, specifically designed for text embedding and ranking tasks. Building upon the dense foundational models of the Qwen3 series, it provides a comprehensive range of text embeddings and reranking models in various sizes (0.6B, 4B, and 8B).

WizardLM-2 7B is the smaller variant of Microsoft AI's latest Wizard model. It is the fastest and achieves comparable performance with existing 10x larger open-source leading models

The Multilingual-E5 models, initialized from XLM-RoBERTa, support up to 512 tokens per input — any longer text will be silently truncated. To ensure optimal performance, always prefix inputs with “query:” or “passage:”, as the model was explicitly trained with this format.

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. 

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.

The Deliberate Model allows for the creation of anything desired, with the potential for better results as the user's knowledge and detail in the prompt increase. The model is ideal for meticulous anatomy artists, creative prompt writers, art designers, and those seeking explicit content.

A sentence transformation model that has been trained on a wide range of datasets, including but not limited to S2ORC, WikiAnwers, PAQ, Stack Exchange, and Yahoo! Answers. Our model can be used for various NLP tasks such as clustering, sentiment analysis, and question answering.

Gemma is a family of lightweight, state-of-the-art open models from Google. Gemma-2-27B delivers the best performance for its size class, and even offers competitive alternatives to models more than twice its size. 

Latest version of the Airoboros model fine-tunned version of llama-2-70b using the Airoboros dataset. This model is currently running jondurbin/airoboros-l2-70b-2.2.1 

Qwen3-Coder-30B-A3B-Instruct is a high-performance code generation model optimized for agentic coding and complex programming tasks. With 30.5B total parameters and 3.3B activated through Mixture-of-Experts architecture, it delivers exceptional efficiency. The model features native support for 256K token context (extendable to 1M), making it ideal for repository-scale code understanding. It excels at tool calling, browser automation, and multi-step coding workflows.

The Llama 90B Vision model is a top-tier, 90-billion-parameter multimodal model designed for the most challenging visual reasoning and language tasks. It offers unparalleled accuracy in image captioning, visual question answering, and advanced image-text comprehension. Pre-trained on vast multimodal datasets and fine-tuned with human feedback, the Llama 90B Vision is engineered to handle the most demanding image-based AI tasks.  This model is perfect for industries requiring cutting-edge multimodal AI capabilities, particularly those dealing with complex, real-time visual and textual analysis.

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. 

Meta developed and released the Meta Llama 3 family of large language models (LLMs), a collection of pretrained and instruction tuned generative text models in 8 and 70B sizes.

The 72 billion parameter Qwen2 excels in language understanding, multilingual capabilities, coding, mathematics, and reasoning.

Whisper is a general-purpose speech recognition model. It is trained on a large dataset of diverse audio and is also a multi-task model that can perform multilingual speech recognition as well as speech translation and language identification.

FLUX.1 [schnell] is a 12 billion parameter rectified flow transformer capable of generating images from text descriptions. This model offers cutting-edge output quality and competitive prompt following, matching the performance of closed source alternatives. Trained using latent adversarial diffusion distillation, FLUX.1 [schnell] can generate high-quality images in only 1 to 4 steps. 

The Qwen3 Embedding model series is the latest proprietary model of the Qwen family, specifically designed for text embedding and ranking tasks. Building upon the dense foundational models of the Qwen3 series, it provides a comprehensive range of text embeddings and reranking models in various sizes (0.6B, 4B, and 8B)

Bria RMBG 2.0 enables seamless removal of backgrounds from images, ideal for professional editing tasks. Trained exclusively on licensed data for safe and risk-free commercial use.

WizardLM-2 8x22B is Microsoft AI's most advanced Wizard model. It demonstrates highly competitive performance compared to those leading proprietary models.

Devstral is an agentic LLM for software engineering tasks. Devstral excels at using tools to explore codebases, editing multiple files and power software engineering agents. 

Janus-Pro is a novel autoregressive framework that unifies multimodal understanding and generation. It addresses the limitations of previous approaches by decoupling visual encoding into separate pathways, while still utilizing a single, unified transformer architecture for processing. The decoupling not only alleviates the conflict between the visual encoder’s roles in understanding and generation, but also enhances the framework’s flexibility. Janus-Pro surpasses previous unified model and matches or exceeds the performance of task-specific models. The simplicity, high flexibility, and effectiveness of Janus-Pro make it a strong candidate for next-generation unified multimodal models.

Bria Erase Foreground precisely removes main subjects or foreground objects from images. Built entirely on licensed data, it is safe and optimized for professional and commercial use.

Llama 2 is a collection of pretrained and fine-tuned generative text models ranging in scale from 7 billion to 70 billion parameters. This is the repository for the 7B fine-tuned model, optimized for dialogue use cases and converted for the Hugging Face Transformers format. 

A Mythomax/MLewd_13B-style merge of selected 70B models  A multi-model merge of several  LLaMA2 70B finetunes for roleplaying and creative work. The goal was to create a model that combines creativity with intelligence for an enhanced experience.

This is the instruction fine-tuned version of Mixtral-8x22B - the latest and largest mixture of experts large language model (LLM) from Mistral AI. This state of the art machine learning model uses a mixture 8 of experts (MoE) 22b models. During inference 2 experts are selected. This architecture allows large models to be fast and cheap at inference.

A LLM-based embedding model with in-context learning capabilities that achieves SOTA performance on BEIR and AIR-Bench. It leverages few-shot examples to enhance task performance.

Over the past few months, we have observed increasingly clear trends toward scaling both total parameters and context lengths in the pursuit of more powerful and agentic artificial intelligence (AI). We are excited to share our latest advancements in addressing these demands, centered on improving scaling efficiency through innovative model architecture. We call this next-generation foundation models Qwen3-Next.

Gemma is an open-source model designed by Google. This is Gemma 1.1 7B (IT), an update over the original instruction-tuned Gemma release. Gemma 1.1 was trained using a novel RLHF method, leading to substantial gains on quality, coding capabilities, factuality, instruction following and multi-turn conversation quality.

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.

We present a sentence transformation model that achieves state-of-the-art results on various NLP tasks without requiring task-specific architectures or fine-tuning. Our approach leverages contrastive learning and utilizes a variety of datasets to learn robust sentence representations. We evaluate our model on several benchmarks and demonstrate its effectiveness in various applications such as text classification, sentiment analysis, named entity recognition, and question answering.

Anthropic’s most powerful model yet and the state-of-the-art coding model. It delivers sustained performance on long-running tasks that require focused effort and thousands of steps, significantly expanding what AI agents can solve. Claude Opus 4 is ideal for powering frontier agent products and features.

Qwen2.5-Coder is the latest series of Code-Specific Qwen large language models (formerly known as CodeQwen). It has significant improvements in code generation, code reasoning and code fixing. A more comprehensive foundation for real-world applications such as Code Agents. Not only enhancing coding capabilities but also maintaining its strengths in mathematics and general competencies.

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.

New model named Chatterbox by Resemble AI's first production-grade open source TTS model. Licensed under MIT, Chatterbox has been benchmarked against leading closed-source systems like ElevenLabs, and is consistently preferred in side-by-side evaluations.  Whether you're working on memes, videos, games, or AI agents, Chatterbox brings your content to life. It's also the first open source TTS model to support emotion exaggeration control, a powerful feature that makes your voices stand out.

The 7 billion parameter Qwen2.5 excels in language understanding, multilingual capabilities, coding, mathematics, and reasoning

You can use cURL or any other http client to run inferences:

```bash
curl -X POST \
    -d '{"input": "The quick brown fox jumps over the lazy dog"}'  \
    -H "Authorization: bearer $DEEPINFRA_TOKEN"  \
    -H 'Content-Type: application/json'  \
    'https://api.deepinfra.com/v1/inference/bosonai/HiggsAudioV2.5'
```

which will give you back something similar to:

```json
{
  "audio": null,
  "input_character_length": 0,
  "output_format": "",
  "words": [
    {
      "end": 1.0,
      "start": 0.0,
      "text": "Hello"
    },
    {
      "end": 5.0,
      "start": 4.0,
      "text": "World"
    }
  ],
  "request_id": null,
  "inference_status": {
    "status": "unknown",
    "runtime_ms": 0,
    "cost": 0.0,
    "tokens_generated": 0,
    "tokens_input": 0
  }
}

```


You can use our command-line tool [deepctl](/docs/advanced/deepctl) to run
inferences:

```bash
deepctl infer \
    -m 'bosonai/HiggsAudioV2.5'  \
    -i 'input=The quick brown fox jumps over the lazy dog'
```

which will give you back something similar to:

```json
{
  "audio": null,
  "input_character_length": 0,
  "output_format": "",
  "words": [
    {
      "end": 1.0,
      "start": 0.0,
      "text": "Hello"
    },
    {
      "end": 5.0,
      "start": 4.0,
      "text": "World"
    }
  ],
  "request_id": null,
  "inference_status": {
    "status": "unknown",
    "runtime_ms": 0,
    "cost": 0.0,
    "tokens_generated": 0,
    "tokens_input": 0
  }
}

```


The DeepInfra OpenAI-compatible Speech API endpoint enables users to effortlessly convert text into speech audio. This document outlines how to integrate and utilize this endpoint to quickly create speech from text inputs, leveraging various audio output formats.

## Create speech

Use the following example of pythong code to generate an audio file from your text input:

```python
from pathlib import Path
from openai import OpenAI

client = OpenAI(base_url="https://api.deepinfra.com/v1/openai",
                api_key="$DEEPINFRA_TOKEN")

speech_file_path = Path(__file__).parent / "speech.mp3"
with client.audio.speech.with_streaming_response.create(
  model="bosonai/HiggsAudioV2.5",
  voice="belinda",
  input="The quick brown fox jumped over the lazy dog.",
  response_format="mp3",
) as response:
  response.stream_to_file(speech_file_path)
```

The API returns the generated audio file in the requested format (e.g., `mp3`, `pcm`). The example above saves the audio output directly as `speech.mp3`.

### Supported Audio Formats

| Format | Description                           | Recommended Usage                          |
|--------|---------------------------------------|--------------------------------------------|
| pcm    | Raw, uncompressed audio               | *Best for lowest latency streaming*        |
| mp3    | Compressed audio suitable for storage | General use and file distribution          |
| opus   | Compressed audio ideal for streaming  | Efficient streaming and real-time playback |
| flac   | Lossless audio compression            | High-quality archival storage              |
| wav    | Uncompressed audio                    | High-quality audio processing tasks        |

### Performance Recommendations

- **For lowest latency (fastest initial audio chunk):**
  - Use the `pcm` output format.

- **For general purposes:**
  - Use the `mp3` or `opus` formats for optimal balance between quality and file size.



The DeepInfra OpenAI-compatible Speech API endpoint enables users to effortlessly convert text into speech audio. This document outlines how to integrate and utilize this endpoint to quickly create speech from text inputs, leveraging various audio output formats.

## Create speech

Use the following example of js code to generate an audio file from your text input:

```javascript
import fs from "fs";
import path from "path";
import OpenAI from "openai";

const openai = new OpenAI(base_url="https://api.deepinfra.com/v1/openai",
                          api_key="$DEEPINFRA_TOKEN");

const speechFile = path.resolve("./speech.mp3");

async function main() {
  const mp3 = await openai.audio.speech.create({
    model: "bosonai/HiggsAudioV2.5",
    voice: "belinda",
    input: "The quick brown fox jumped over the lazy dog.",
    response_format: "mp3",
  });
  console.log(speechFile);
  const buffer = Buffer.from(await mp3.arrayBuffer());
  await fs.promises.writeFile(speechFile, buffer);
}
main();
```

The API returns the generated audio file in the requested format (e.g., `mp3`, `pcm`). The example above saves the audio output directly as `speech.mp3`.

### Supported Audio Formats

| Format | Description                           | Recommended Usage                          |
|--------|---------------------------------------|--------------------------------------------|
| pcm    | Raw, uncompressed audio               | *Best for lowest latency streaming*        |
| mp3    | Compressed audio suitable for storage | General use and file distribution          |
| opus   | Compressed audio ideal for streaming  | Efficient streaming and real-time playback |
| flac   | Lossless audio compression            | High-quality archival storage              |
| wav    | Uncompressed audio                    | High-quality audio processing tasks        |

### Performance Recommendations

- **For lowest latency (fastest initial audio chunk):**
  - Use the `pcm` output format.

- **For general purposes:**
  - Use the `mp3` or `opus` formats for optimal balance between quality and file size.



The DeepInfra OpenAI-compatible Speech API endpoint enables users to effortlessly convert text into speech audio. This document outlines how to integrate and utilize this endpoint to quickly create speech from text inputs, leveraging various audio output formats.

## Create speech

Use the following example `curl` request to generate an audio file from your text input:

```bash
curl https://api.deepinfra.com/v1/openai/audio/speech \
  -H "Authorization: Bearer $DEEPINFRA_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "bosonai/HiggsAudioV2.5",
    "input": "The quick brown fox jumped over the lazy dog.",
    "voice": "belinda",
    "response_format": "mp3"
  }' \
  --output speech.mp3
```

The API returns the generated audio file in the requested format (e.g., `mp3`, `pcm`). The example above saves the audio output directly as `speech.mp3`.

### Supported Audio Formats

| Format | Description                           | Recommended Usage                          |
|--------|---------------------------------------|--------------------------------------------|
| pcm    | Raw, uncompressed audio               | *Best for lowest latency streaming*        |
| mp3    | Compressed audio suitable for storage | General use and file distribution          |
| opus   | Compressed audio ideal for streaming  | Efficient streaming and real-time playback |
| flac   | Lossless audio compression            | High-quality archival storage              |
| wav    | Uncompressed audio                    | High-quality audio processing tasks        |

### Performance Recommendations

- **For lowest latency (fastest initial audio chunk):**
  - Use the `pcm` output format.

- **For general purposes:**
  - Use the `mp3` or `opus` formats for optimal balance between quality and file size.



The DeepInfra ElevenLabs-compatible Speech API endpoint allows users to seamlessly convert text inputs into high-quality speech audio. This document provides guidance on integrating and using this endpoint to generate realistic audio files efficiently.

Use the following examples of `py` request to generate an audio file from your text input:

## Create Speech (non-streaming)

```bash
from elevenlabs import ElevenLabs

client = ElevenLabs(
    api_key="$DEEPINFRA_TOKEN",
    base_url="https://api.deepinfra.com/",
)
client.text_to_speech.convert(
    voice_id="belinda",
    output_format="mp3",
    text="The quick brown fox jumped over the lazy dog.",
    model_id="bosonai/HiggsAudioV2.5",
)
```

## Create Speech with Streaming

```bash
from elevenlabs import ElevenLabs

client = ElevenLabs(
    api_key="$DEEPINFRA_TOKEN",
)
client.text_to_speech.convert_as_stream(
    voice_id="belinda",
    output_format="pcm",
    text="The quick brown fox jumped over the lazy dog.",
    model_id="bosonai/HiggsAudioV2.5",
)
```

The API returns the generated audio in the requested format, such as `mp3` or `pcm`. The example above saves the audio directly as `speech.mp3`.

### Supported Audio Formats

| Format | Description                           | Recommended Usage                          |
|--------|---------------------------------------|--------------------------------------------|
| pcm    | Raw, uncompressed audio               | *Lowest latency streaming scenarios*       |
| mp3    | Compressed audio suitable for storage | General use and easy file sharing          |
| opus   | Compressed audio ideal for streaming  | Real-time audio streaming applications     |
| flac   | Lossless audio compression            | Archival storage and high-fidelity needs   |
| wav    | Uncompressed audio                    | Audio processing and editing tasks         |

### Performance Recommendations

- **Lowest latency:** Using the `pcm` output format for streaming applications is **HIGHLY RECOMMENDED**.
- **General use:** Use `mp3` or `opus` formats for the best trade-off between quality and file size.



The DeepInfra ElevenLabs-compatible Speech API endpoint allows users to seamlessly convert text inputs into high-quality speech audio. This document provides guidance on integrating and using this endpoint to generate realistic audio files efficiently.

Use the following examples of `js` request to generate an audio file from your text input:

## Create Speech (non-streaming)

```bash
import { ElevenLabsClient } from "elevenlabs";

const client = new ElevenLabsClient({ apiKey: "$DEEPINFRA_TOKEN", base_url: "https://api.deepinfra.com/" });
await client.textToSpeech.convert("belinda", {
    output_format: "mp3",
    text: "The quick brown fox jumped over the lazy dog.",
    model_id: "bosonai/HiggsAudioV2.5"
});
```

## Create Speech with Streaming

```bash
import { ElevenLabsClient } from "elevenlabs";

const client = new ElevenLabsClient({ apiKey: "$DEEPINFRA_TOKEN" });
await client.textToSpeech.convert("belinda", {
    output_format: "pcm",
    text: "The quick brown fox jumped over the lazy dog.",
    model_id: "bosonai/HiggsAudioV2.5"
});
```

The API returns the generated audio in the requested format, such as `mp3` or `pcm`. The example above saves the audio directly as `speech.mp3`.

### Supported Audio Formats

| Format | Description                           | Recommended Usage                          |
|--------|---------------------------------------|--------------------------------------------|
| pcm    | Raw, uncompressed audio               | *Lowest latency streaming scenarios*       |
| mp3    | Compressed audio suitable for storage | General use and easy file sharing          |
| opus   | Compressed audio ideal for streaming  | Real-time audio streaming applications     |
| flac   | Lossless audio compression            | Archival storage and high-fidelity needs   |
| wav    | Uncompressed audio                    | Audio processing and editing tasks         |

### Performance Recommendations

- **Lowest latency:** Using the `pcm` output format for streaming applications is **HIGHLY RECOMMENDED**.
- **General use:** Use `mp3` or `opus` formats for the best trade-off between quality and file size.



The DeepInfra ElevenLabs-compatible Speech API endpoint allows users to seamlessly convert text inputs into high-quality speech audio. This document provides guidance on integrating and using this endpoint to generate realistic audio files efficiently.

Use the following examples of `curl` request to generate an audio file from your text input:

## Create Speech (non-streaming)

```bash
curl -X POST "https://api.deepinfra.com/v1/text-to-speech/belinda" \
     -H "xi-api-key: $DEEPINFRA_TOKEN" \
     -H "Content-Type: application/json" \
     -d '{
  "text": "The quick brown fox jumped over the lazy dog.",
  "model_id": "bosonai/HiggsAudioV2.5",
  "output_format": "mp3",
}' --output speech.mp3
```

## Create Speech with Streaming

```bash
curl -X POST "https://api.deepinfra.com/v1/text-to-speech/belinda/stream" \
     -H "xi-api-key: $DEEPINFRA_TOKEN" \
     -H "Content-Type: application/json" \
     -d '{
  "text": "The quick brown fox jumped over the lazy dog.",
  "model_id": "bosonai/HiggsAudioV2.5",
  "output_format": "pcm",
}' --output speech.pcm
```

The API returns the generated audio in the requested format, such as `mp3` or `pcm`. The example above saves the audio directly as `speech.mp3`.

### Supported Audio Formats

| Format | Description                           | Recommended Usage                          |
|--------|---------------------------------------|--------------------------------------------|
| pcm    | Raw, uncompressed audio               | *Lowest latency streaming scenarios*       |
| mp3    | Compressed audio suitable for storage | General use and easy file sharing          |
| opus   | Compressed audio ideal for streaming  | Real-time audio streaming applications     |
| flac   | Lossless audio compression            | Archival storage and high-fidelity needs   |
| wav    | Uncompressed audio                    | Audio processing and editing tasks         |

### Performance Recommendations

- **Lowest latency:** Using the `pcm` output format for streaming applications is **HIGHLY RECOMMENDED**.
- **General use:** Use `mp3` or `opus` formats for the best trade-off between quality and file size.



input

Voice name to use. Can be a preset voice (belinda, chadwick, etc.) or a custom voice ID for voice cloning.

voice

response_format

stream

The webhook to call when inference is done, by default you will get the output in the response of your inference request

webhook

Select the desired voice for the speech output.

HiggsAudioTtsVoice

Input text

Response Format

Stream

Voice

Webhook

HiggsAudioTextToSpeechIn

estimated cost billed for the request in USD

Cost

Runtime Ms

Status

Tokens Generated

Tokens Input

InferenceReplyStatus

Start

Text

Word

Audio

Object containing the status of the inference request

Input Character Length

Output Format

Request Id

Words

TextToSpeechOut

The service tier used for processing the request. When set to 'priority', the request will be processed with higher priority (only applies to models that support it).

service_tier

model

speed

extra_body

ServiceTier

Select the desired format for the speech output. Supported formats include mp3, opus, flac, wav, and pcm.