Best AI Models in the USA - Page 7
View:
Compare the Top AI Models in the USA as of February 2026 - Page 7
Sort By:
-
1
EVI 3
Hume AI
Hume AI's EVI 3 is a third-generation speech-language model that streams in user speech and forms natural, expressive speech and language responses. At conversational latency, it produces the same quality of speech as our text-to-speech model, Octave. Simultaneously, it responds with the same intelligence as the most advanced LLMs of similar latency. It also communicates with reasoning models and web search systems as it speaks, “thinking fast and slow” to match the intelligence of any frontier AI system. EVI 3 can instantly generate new voices and personalities instead of being limited to a handful of speakers. For instance, users can speak to any of the more than 100,000 custom voices already created on our text-to-speech platform, each with an inferred personality. No matter the voice, it responds with a wide range of emotions or styles, implicitly or on command.Starting Price: Free -
2
HunyuanVideo-Avatar
Tencent-Hunyuan
HunyuanVideo‑Avatar supports animating any input avatar images to high‑dynamic, emotion‑controllable videos using simple audio conditions. It is a multimodal diffusion transformer (MM‑DiT)‑based model capable of generating dynamic, emotion‑controllable, multi‑character dialogue videos. It accepts multi‑style avatar inputs, photorealistic, cartoon, 3D‑rendered, anthropomorphic, at arbitrary scales from portrait to full body. Provides a character image injection module that ensures strong character consistency while enabling dynamic motion; an Audio Emotion Module (AEM) that extracts emotional cues from a reference image to enable fine‑grained emotion control over generated video; and a Face‑Aware Audio Adapter (FAA) that isolates audio influence to specific face regions via latent‑level masking, supporting independent audio‑driven animation in multi‑character scenarios.Starting Price: Free -
3
Act-Two
Runway AI
Act-Two enables animation of any character by transferring movements, expressions, and speech from a driving performance video onto a static image or reference video of your character. By selecting the Gen‑4 Video model and then the Act‑Two icon in Runway’s web interface, you supply two inputs; a performance video of an actor enacting your desired scene and a character input (either a single image or a video clip), and optionally enable gesture control to map hand and body movements onto character images. Act‑Two automatically adds environmental and camera motion to still images, supports a range of angles, non‑human subjects, and artistic styles, and retains original scene dynamics when using character videos (though with facial rather than full‑body gesture mapping). Users can adjust facial expressiveness on a sliding scale to balance natural motion with character consistency, preview results in real time, and generate high‑resolution clips up to 30 seconds long.Starting Price: $12 per month -
4
Decart Mirage
Decart Mirage
Mirage is the world’s first real‑time, autoregressive video‑to‑video transformation model that instantly turns any live video, game, or camera feed into a new digital world without pre‑rendering. Powered by Live‑Stream Diffusion (LSD) technology, it processes inputs at 24 FPS with under 40 ms latency, ensuring smooth, continuous transformations while preserving motion and structure. Mirage supports universal input, webcams, gameplay, movies, and live streams, and applies text‑prompted style changes on the fly. Its advanced history‑augmentation mechanism maintains temporal coherence across frames, avoiding the glitches common in diffusion‑only approaches. GPU‑accelerated custom CUDA kernels deliver up to 16× faster performance than traditional methods, enabling infinite streaming without interruption. It offers real‑time mobile and desktop previews, seamless integration with any video source, and flexible deployment.Starting Price: Free -
5
Qwen3-Coder
Qwen
Qwen3‑Coder is an agentic code model available in multiple sizes, led by the 480B‑parameter Mixture‑of‑Experts variant (35B active) that natively supports 256K‑token contexts (extendable to 1M) and achieves state‑of‑the‑art results comparable to Claude Sonnet 4. Pre‑training on 7.5T tokens (70 % code) and synthetic data cleaned via Qwen2.5‑Coder optimized both coding proficiency and general abilities, while post‑training employs large‑scale, execution‑driven reinforcement learning, scaling test‑case generation for diverse coding challenges, and long‑horizon RL across 20,000 parallel environments to excel on multi‑turn software‑engineering benchmarks like SWE‑Bench Verified without test‑time scaling. Alongside the model, the open source Qwen Code CLI (forked from Gemini Code) unleashes Qwen3‑Coder in agentic workflows with customized prompts, function calling protocols, and seamless integration with Node.js, OpenAI SDKs, and environment variables.Starting Price: Free -
6
GLM-4.5-Air
Z.ai
Z.ai is a free AI assistant that brings presentations, writing, and coding into one conversational interface. Leveraging large language models, it lets you generate polished slide decks with AI slides, craft professional‑grade text for emails, reports, or blogs, and write or debug complex code. Beyond content creation, Z.ai supports deep research and information search, helping you gather facts, summarize long documents, and overcome writer’s block, while its code agent can explain snippets, refactor functions, or build scripts from scratch. An intuitive chat interface means no steep learning curves: simply tell Z.ai what you need, a strategic deck, marketing copy, or a data‑analysis script, and get instant, contextually relevant results. With support for multiple languages (including Chinese), native function calling, and up to 128K token context length, Z.ai handles everything from brainstorming ideas to automating repetitive writing or coding.Starting Price: Free -
7
ByteDance Seed
ByteDance
Seed Diffusion Preview is a large-scale, code-focused language model that uses discrete-state diffusion to generate code non-sequentially, achieving dramatically faster inference without sacrificing quality by decoupling generation from the token-by-token bottleneck of autoregressive models. It combines a two-stage curriculum, mask-based corruption followed by edit-based augmentation, to robustly train a standard dense Transformer, striking a balance between speed and accuracy and avoiding shortcuts like carry-over unmasking to preserve principled density estimation. The model delivers an inference speed of 2,146 tokens/sec on H20 GPUs, outperforming contemporary diffusion baselines while matching or exceeding their accuracy on standard code benchmarks, including editing tasks, thereby establishing a new speed-quality Pareto frontier and demonstrating discrete diffusion’s practical viability for real-world code generation.Starting Price: Free -
8
Qwen-Image
Alibaba
Qwen-Image is a multimodal diffusion transformer (MMDiT) foundation model offering state-of-the-art image generation, text rendering, editing, and understanding. It excels at complex text integration, seamlessly embedding alphabetic and logographic scripts into visuals with typographic fidelity, and supports diverse artistic styles from photorealism to impressionism, anime, and minimalist design. Beyond creation, it enables advanced image editing operations such as style transfer, object insertion or removal, detail enhancement, in-image text editing, and human pose manipulation through intuitive prompts. Its built-in vision understanding tasks, including object detection, semantic segmentation, depth and edge estimation, novel view synthesis, and super-resolution, extend its capabilities into intelligent visual comprehension. Qwen-Image is accessible via popular libraries like Hugging Face Diffusers and integrates prompt-enhancement tools for multilingual support.Starting Price: Free -
9
FLUX.1 Krea
Krea
FLUX.1 Krea is an open source, guidance-distilled 12 billion-parameter diffusion transformer released by Krea in collaboration with Black Forest Labs, engineered to deliver superior aesthetic control and photorealism while eschewing the generic “AI look.” Fully compatible with the FLUX.1-dev ecosystem, it starts from a raw, untainted base model (flux-dev-raw) rich in world knowledge and employs a two-phase post-training pipeline, supervised fine-tuning on a hand-curated mix of high-quality and synthetic samples, followed by reinforcement learning from human feedback using opinionated preference data, to bias outputs toward a distinct style. By leveraging negative prompts during pre-training, custom loss functions for classifier-free guidance, and targeted preference labels, it achieves significant quality improvements with under one million examples, all without extensive prompting or additional LoRA modules.Starting Price: Free -
10
GPT-5 mini
OpenAI
GPT-5 mini is a streamlined, faster, and more affordable variant of OpenAI’s GPT-5, optimized for well-defined tasks and precise prompts. It supports text and image inputs and delivers high-quality text outputs with a 400,000-token context window and up to 128,000 output tokens. This model excels at rapid response times, making it suitable for applications requiring fast, accurate language understanding without the full overhead of GPT-5. Pricing is cost-effective, with input tokens at $0.25 per million and output tokens at $2 per million, providing savings over the flagship model. GPT-5 mini supports advanced features like streaming, function calling, structured outputs, and fine-tuning, but does not support audio input or image generation. It integrates well with various API endpoints including chat completions, responses, and embeddings, making it versatile for many AI-powered tasks.Starting Price: $0.25 per 1M tokens -
11
GPT-5 nano
OpenAI
GPT-5 nano is OpenAI’s fastest and most affordable version of the GPT-5 family, designed for high-speed text processing tasks like summarization and classification. It supports text and image inputs, generating high-quality text outputs with a large 400,000-token context window and up to 128,000 output tokens. GPT-5 nano offers very fast response times, making it ideal for applications requiring quick turnaround without sacrificing quality. Pricing is extremely competitive, with input tokens costing $0.05 per million and output tokens $0.40 per million, making it accessible for budget-conscious projects. The model supports advanced API features such as streaming, function calling, structured outputs, and fine-tuning. While it supports image input, it does not handle audio input or web search, focusing on core text tasks efficiently.Starting Price: $0.05 per 1M tokens -
12
NVIDIA Cosmos
NVIDIA
NVIDIA Cosmos is a developer-first platform of state-of-the-art generative World Foundation Models (WFMs), advanced video tokenizers, guardrails, and an accelerated data processing and curation pipeline designed to supercharge physical AI development. It enables developers working on autonomous vehicles, robotics, and video analytics AI agents to generate photorealistic, physics-aware synthetic video data, trained on an immense dataset including 20 million hours of real-world and simulated video, to rapidly simulate future scenarios, train world models, and fine‑tune custom behaviors. It includes three core WFM types; Cosmos Predict, capable of generating up to 30 seconds of continuous video from multimodal inputs; Cosmos Transfer, which adapts simulations across environments and lighting for versatile domain augmentation; and Cosmos Reason, a vision-language model that applies structured reasoning to interpret spatial-temporal data for planning and decision-making.Starting Price: Free -
13
NVIDIA Isaac GR00T
NVIDIA
NVIDIA Isaac GR00T (Generalist Robot 00 Technology) is a research-driven platform for developing general-purpose humanoid robot foundation models and data pipelines. It includes models like Isaac GR00T-N, and synthetic motion blueprints, GR00T-Mimic for augmenting demonstrations, and GR00T-Dreams for generating novel synthetic trajectories, to accelerate humanoid robotics development. Recently, the open source Isaac GR00T N1 foundation model debuted, featuring a dual-system cognitive architecture, a fast-reacting “System 1” action model, and a deliberative, language-enabled “System 2” reasoning model. The updated GR00T N1.5 introduces enhancements such as improved vision-language grounding, better language command following, few-shot adaptability, and new robot embodiment support. Together with tools like Isaac Sim, Lab, and Omniverse, GR00T empowers developers to train, simulate, post-train, and deploy adaptable humanoid agents using both real and synthetic data.Starting Price: Free -
14
DeepSeek V3.1
DeepSeek
DeepSeek V3.1 is a groundbreaking open-weight large language model featuring a massive 685-billion parameters and an extended 128,000‑token context window, enabling it to process documents equivalent to 400-page books in a single prompt. It delivers integrated capabilities for chat, reasoning, and code generation within a unified hybrid architecture, seamlessly blending these functions into one coherent model. V3.1 supports a variety of tensor formats to give developers flexibility in optimizing performance across different hardware. Early benchmark results show robust performance, including a 71.6% score on the Aider coding benchmark, putting it on par with or ahead of systems like Claude Opus 4 and doing so at a far lower cost. Made available under an open source license on Hugging Face with minimal fanfare, DeepSeek V3.1 is poised to reshape access to high-performance AI, challenging traditional proprietary models.Starting Price: Free -
15
gpt-realtime
OpenAI
GPT-Realtime is OpenAI’s most advanced, production-ready speech-to-speech model, now accessible through the fully available Realtime API. It delivers remarkably natural, expressive audio with fine-grained control over tone, pace, and accent. The model can comprehend nuanced human audio, including laughter, switch languages mid-sentence, and accurately process alphanumeric details like phone numbers across multiple languages. It significantly improves reasoning and instruction-following (achieving 82.8% on the BigBench Audio benchmark and 30.5% on MultiChallenge) and boasts enhanced function calling, now more reliable, timely, and accurate (scoring 66.5% on ComplexFuncBench). The model supports asynchronous tool invocation so conversations remain fluid even during long-running calls. The Realtime API also offers innovative capabilities such as image input support, SIP phone network integration, remote MCP server connection, and reusable conversation prompts.Starting Price: $20 per month -
16
Hermes 4
Nous Research
Hermes 4 is the latest evolution in Nous Research’s line of neutrally aligned, steerable foundational models, featuring novel hybrid reasoners that can dynamically shift between expressive, creative responses and efficient, standard replies based on user prompts. The model is designed to respond to system and user instructions, rather than adhering to any corporate ethics framework, producing interactions that feel more humanistic, less lecturing or sycophantic, and encouraging roleplay and creativity. By incorporating a special tag in prompts, users can trigger deeper, internally token-intensive reasoning when tackling complex problems, while retaining prompt efficiency when such depth isn't required. Trained on a dataset 50 times larger than that of Hermes 3, much of which was synthetically generated using Atropos, Hermes 4 shows significant performance improvements.Starting Price: Free -
17
K2 Think
Institute of Foundation Models
K2 Think is an open source advanced reasoning model developed collaboratively by the Institute of Foundation Models at MBZUAI and G42. Despite only having 32 billion parameters, it delivers performance comparable to flagship models with many more parameters. It excels in mathematical reasoning, achieving top scores on competitive benchmarks such as AIME ’24/’25, HMMT ’25, and OMNI-Math-HARD. K2 Think is part of a suite of UAE-developed open models, alongside Jais (Arabic), NANDA (Hindi), and SHERKALA (Kazakh), and builds on the foundation laid by K2-65B, the fully reproducible open source foundation model released in 2024. The model is designed to be open, fast, and flexible, offering a web app interface for exploration, and with its efficiency in parameter positioning, it is a breakthrough in compact architectures for advanced AI reasoning.Starting Price: Free -
18
Ray3
Luma AI
Ray3 is an advanced video generation model by Luma Labs, built to help creators tell richer visual stories with pro-level fidelity. It introduces native 16-bit High Dynamic Range (HDR) video generations, enabling more vibrant color, deeper contrasts, and overall pro studio pipelines. The model incorporates sophisticated physics and improved consistency (motion, anatomy, lighting, reflections), supports visual controls, and has a draft mode that lets you explore ideas quickly before up-rendering selected pieces into high-fidelity 4K HDR output. Ray3 can interpret prompts with nuance, reason about intent, self-evaluate early drafts, and adjust to satisfy the articulation of scene and motion more accurately. Other features include support for keyframes, loop and extend functions, upscaling, and export of frames for seamless integration into professional workflows.Starting Price: $9.99 per month -
19
DeepSeek-V3.1-Terminus
DeepSeek
DeepSeek has released DeepSeek-V3.1-Terminus, which enhances the V3.1 architecture by incorporating user feedback to improve output stability, consistency, and agent performance. It notably reduces instances of mixed Chinese/English character output and unintended garbled characters, resulting in cleaner, more consistent language generation. The update upgrades both the code agent and search agent subsystems to yield stronger, more reliable performance across benchmarks. DeepSeek-V3.1-Terminus is also available as an open source model, and its weights are published on Hugging Face. The model structure remains the same as DeepSeek-V3, ensuring compatibility with existing deployment methods, with updated inference demos provided for community use. While trained at a scale of 685B parameters, the model includes FP8, BF16, and F32 tensor formats, offering flexibility across environments.Starting Price: Free -
20
Qwen3-Max
Alibaba
Qwen3-Max is Alibaba’s latest trillion-parameter large language model, designed to push performance in agentic tasks, coding, reasoning, and long-context processing. It is built atop the Qwen3 family and benefits from the architectural, training, and inference advances introduced there; mixing thinker and non-thinker modes, a “thinking budget” mechanism, and support for dynamic mode switching based on complexity. The model reportedly processes extremely long inputs (hundreds of thousands of tokens), supports tool invocation, and exhibits strong performance on benchmarks in coding, multi-step reasoning, and agent benchmarks (e.g., Tau2-Bench). While its initial variant emphasizes instruction following (non-thinking mode), Alibaba plans to bring reasoning capabilities online to enable autonomous agent behavior. Qwen3-Max inherits multilingual support and extensive pretraining on trillions of tokens, and it is delivered via API interfaces compatible with OpenAI-style functions.Starting Price: Free -
21
GLM-4.6
Zhipu AI
GLM-4.6 advances upon its predecessor with stronger reasoning, coding, and agentic capabilities: it demonstrates clear improvements in inferential performance, supports tool use during inference, and more effectively integrates into agent frameworks. In benchmark tests spanning reasoning, coding, and agents, GLM-4.6 outperforms GLM-4.5 and shows competitive strength against models such as DeepSeek-V3.2-Exp and Claude Sonnet 4, though it still trails Claude Sonnet 4.5 in pure coding performance. In real-world tests using an extended “CC-Bench” suite across front-end development, tool building, data analysis, and algorithmic tasks, GLM-4.6 beats GLM-4.5 and approaches parity with Claude Sonnet 4, winning ~48.6% of head-to-head comparisons, while also achieving ~15% better token efficiency. GLM-4.6 is available via the Z.ai API, and developers can integrate it as an LLM backend or agent core using the platform’s API.Starting Price: Free -
22
DeepSeek-V3.2-Exp
DeepSeek
Introducing DeepSeek-V3.2-Exp, our latest experimental model built on V3.1-Terminus, debuting DeepSeek Sparse Attention (DSA) for faster and more efficient inference and training on long contexts. DSA enables fine-grained sparse attention with minimal loss in output quality, boosting performance for long-context tasks while reducing compute costs. Benchmarks indicate that V3.2-Exp performs on par with V3.1-Terminus despite these efficiency gains. The model is now live across app, web, and API. Alongside this, the DeepSeek API prices have been cut by over 50% immediately to make access more affordable. For a transitional period, users can still access V3.1-Terminus via a temporary API endpoint until October 15, 2025. DeepSeek welcomes feedback on DSA via its feedback portal. In conjunction with the release, DeepSeek-V3.2-Exp has been open-sourced: the model weights and supporting technology (including key GPU kernels in TileLang and CUDA) are available on Hugging Face.Starting Price: Free -
23
gpt-4o-mini Realtime
OpenAI
The gpt-4o-mini-realtime-preview model is a compact, lower-cost, realtime variant of GPT-4o designed to power speech and text interactions with low latency. It supports both text and audio inputs and outputs, enabling “speech in, speech out” conversational experiences via a persistent WebSocket or WebRTC connection. Unlike larger GPT-4o models, it currently does not support image or structured output modalities, focusing strictly on real-time voice/text use cases. Developers can open a real-time session via the /realtime/sessions endpoint to obtain an ephemeral key, then stream user audio (or text) and receive responses in real time over the same connection. The model is part of the early preview family (version 2024-12-17), intended primarily for testing and feedback rather than full production loads. Usage is subject to rate limits and may evolve during the preview period. Because it is multimodal in audio/text only, it enables use cases such as conversational voice agents.Starting Price: $0.60 per input -
24
Hunyuan-Vision-1.5
Tencent
HunyuanVision is a cutting-edge vision-language model developed by Tencent’s Hunyuan team. It uses a mamba-transformer hybrid architecture to deliver strong performance and efficient inference in multimodal reasoning tasks. The version Hunyuan-Vision-1.5 is designed for “thinking on images,” meaning it not only understands vision+language content, but can perform deeper reasoning that involves manipulating or reflecting on image inputs, such as cropping, zooming, pointing, box drawing, or drawing on the image to acquire additional knowledge. It supports a variety of vision tasks (image + video recognition, OCR, diagram understanding), visual reasoning, and even 3D spatial comprehension, all in a unified multilingual framework. The model is built to work seamlessly across languages and tasks and is intended to be open sourced (including checkpoints, technical report, inference support) to encourage the community to experiment and adopt.Starting Price: Free -
25
Gemini Enterprise
Google
Gemini Enterprise is a comprehensive AI platform built by Google Cloud designed to bring the full power of Google’s advanced AI models, agent-creation tools, and enterprise-grade data access into everyday workflows. The solution offers a unified chat interface that lets employees interact with internal documents, applications, data sources, and custom AI agents. At its core, Gemini Enterprise comprises six key components: the Gemini family of large multimodal models, an agent orchestration workbench (formerly Google Agentspace), pre-built starter agents, robust data-integration connectors to business systems, extensive security and governance controls, and a partner ecosystem for tailored integrations. It is engineered to scale across departments and enterprises, enabling users to build no-code or low-code agents that automate tasks, such as research synthesis, customer support response, code assist, contract analysis, and more, while operating within corporate compliance standards.Starting Price: $21 per month -
26
Claude Haiku 4.5
Anthropic
Anthropic has launched Claude Haiku 4.5, its latest small-language model designed to deliver near-frontier performance at significantly lower cost. The model provides similar coding and reasoning quality as the company’s mid-tier Sonnet 4, yet it runs at roughly one-third of the cost and more than twice the speed. In benchmarks cited by Anthropic, Haiku 4.5 meets or exceeds Sonnet 4’s performance in key tasks such as code generation and multi-step “computer use” workflows. It is optimized for real-time, low-latency scenarios such as chat assistants, customer service agents, and pair-programming support. Haiku 4.5 is made available via the Claude API under the identifier “claude-haiku-4-5” and supports large-scale deployments where cost, responsiveness, and near-frontier intelligence matter. Claude Haiku 4.5 is available now on Claude Code and our apps. Its efficiency means you can accomplish more within your usage limits while maintaining premium model performance.Starting Price: $1 per million input tokens -
27
MiniMax M2
MiniMax
MiniMax M2 is an open source foundation model built specifically for agentic applications and coding workflows, striking a new balance of performance, speed, and cost. It excels in end-to-end development scenarios, handling programming, tool-calling, and complex, long-chain workflows with capabilities such as Python integration, while delivering inference speeds of around 100 tokens per second and offering API pricing at just ~8% of the cost of comparable proprietary models. The model supports “Lightning Mode” for high-speed, lightweight agent tasks, and “Pro Mode” for in-depth full-stack development, report generation, and web-based tool orchestration; its weights are fully open source and available for local deployment with vLLM or SGLang. MiniMax M2 positions itself as a production-ready model that enables agents to complete independent tasks, such as data analysis, programming, tool orchestration, and large-scale multi-step logic at real organizational scale.Starting Price: $0.30 per million input tokens -
28
GPT-5.1-Codex
OpenAI
GPT-5.1-Codex is a specialized version of the GPT-5.1 model built for software engineering and agentic coding workflows. It is optimized for both interactive development sessions and long-horizon, autonomous execution of complex engineering tasks, such as building projects from scratch, developing features, debugging, performing large-scale refactoring, and code review. It supports tool-use, integrates naturally with developer environments, and adapts reasoning effort dynamically, moving quickly on simple tasks while spending more time on deep ones. The model is described as producing cleaner and higher-quality code outputs compared to general models, with closer adherence to developer instructions and fewer hallucinations. GPT-5.1-Codex is available via the Responses API route (rather than a standard chat API) and comes in variants including “mini” for cost-sensitive usage and “max” for the highest capability.Starting Price: $1.25 per input -
29
SAM 3D
Meta
SAM 3D is a pair of advanced foundation models designed to convert a single standard RGB image into a high-fidelity 3D reconstruction of either objects or human bodies. It comprises SAM 3D Objects, which recovers full 3D geometry, texture, and layout of objects within real-world scenes, handling clutter, occlusions, and diverse lighting, and SAM 3D Body, which produces animatable human mesh models with detailed pose and shape, built on the “Meta Momentum Human Rig” (MHR) format. It is engineered to generalize across in-the-wild images without further training or finetuning: you upload an image, prompt the model by selecting the object or person, and it outputs a downloadable asset ready for use in 3D applications. SAM 3D emphasizes open vocabulary reconstruction (any object category), multi-view consistency, occlusion reasoning, and a massive new dataset of over one million annotated real-world images, enabling its robustness.Starting Price: Free -
30
Olmo 3
Ai2
Olmo 3 is a fully open model family spanning 7 billion and 32 billion parameter variants that delivers not only high-performing base, reasoning, instruction, and reinforcement-learning models, but also exposure of the entire model flow, including raw training data, intermediate checkpoints, training code, long-context support (65,536 token window), and provenance tooling. Starting with the Dolma 3 dataset (≈9 trillion tokens) and its disciplined mix of web text, scientific PDFs, code, and long-form documents, the pre-training, mid-training, and long-context phases shape the base models, which are then post-trained via supervised fine-tuning, direct preference optimisation, and RL with verifiable rewards to yield the Think and Instruct variants. The 32 B Think model is described as the strongest fully open reasoning model to date, competitively close to closed-weight peers in math, code, and complex reasoning.Starting Price: Free
