Alternatives to Ai2 OLMoE
Compare Ai2 OLMoE alternatives for your business or organization using the curated list below. SourceForge ranks the best alternatives to Ai2 OLMoE in 2026. Compare features, ratings, user reviews, pricing, and more from Ai2 OLMoE competitors and alternatives in order to make an informed decision for your business.
-
1
MAI-1-preview
Microsoft
MAI-1 Preview is Microsoft AI’s first end-to-end trained foundation model, built entirely in-house as a mixture-of-experts architecture. Pre-trained and post-trained on approximately 15,000 NVIDIA H100 GPUs, it is designed to follow instructions and generate helpful, responsive text for everyday user queries, representing a prototype of future Copilot capabilities. Now available for public testing on LMArena, MAI-1 Preview delivers an early glimpse into the platform’s trajectory, with plans to roll out select text-based applications within Copilot over the coming weeks to gather user feedback and refine performance. Microsoft reinforces that it will continue combining its own models, partner models, and developments from the open-source community to flexibly power experiences across millions of unique interactions each day. -
2
Mistral Small 4
Mistral AI
Mistral Small 4 is an advanced open-source AI model developed by Mistral AI that combines reasoning, coding, and multimodal capabilities into a single system. It unifies the strengths of previous models such as Magistral for reasoning, Pixtral for multimodal processing, and Devstral for agentic coding tasks. The model can handle both text and image inputs, allowing it to perform tasks ranging from conversational chat to visual analysis and document understanding. Built with a mixture-of-experts architecture, Mistral Small 4 delivers efficient performance while scaling to complex workloads. It also features a configurable reasoning parameter that allows users to switch between fast responses and deeper analytical outputs. With a large context window and optimized inference performance, the model supports long-form interactions and complex workflows.Starting Price: Free -
3
Qwen2
Alibaba
Qwen2 is the large language model series developed by Qwen team, Alibaba Cloud. Qwen2 is a series of large language models developed by the Qwen team at Alibaba Cloud. It includes both base language models and instruction-tuned models, ranging from 0.5 billion to 72 billion parameters, and features both dense models and a Mixture-of-Experts model. The Qwen2 series is designed to surpass most previous open-weight models, including its predecessor Qwen1.5, and to compete with proprietary models across a broad spectrum of benchmarks in language understanding, generation, multilingual capabilities, coding, mathematics, and reasoning.Starting Price: Free -
4
GigaChat 3 Ultra
Sberbank
GigaChat 3 Ultra is a 702-billion-parameter Mixture-of-Experts model built from scratch to deliver frontier-level reasoning, multilingual capability, and deep Russian-language fluency. It activates just 36 billion parameters per token, enabling massive scale with practical inference speeds. The model was trained on a 14-trillion-token corpus combining natural, multilingual, and high-quality synthetic data to strengthen reasoning, math, coding, and linguistic performance. Unlike modified foreign checkpoints, GigaChat 3 Ultra is entirely original—giving developers full control, modern alignment, and a dataset free of inherited limitations. Its architecture leverages MoE, MTP, and MLA to match open-source ecosystems and integrate easily with popular inference and fine-tuning tools. With leading results on Russian benchmarks and competitive performance on global tasks, GigaChat 3 Ultra represents one of the largest and most capable open-source LLMs in the world.Starting Price: Free -
5
Ministral 3
Mistral AI
Mistral 3 is the latest generation of open-weight AI models from Mistral AI, offering a full family of models, from small, edge-optimized versions to a flagship, large-scale multimodal model. The lineup includes three compact “Ministral 3” models (3B, 8B, and 14B parameters) designed for efficiency and deployment on constrained hardware (even laptops, drones, or edge devices), plus the powerful “Mistral Large 3,” a sparse mixture-of-experts model with 675 billion total parameters (41 billion active). The models support multimodal and multilingual tasks, not only text, but also image understanding, and have demonstrated best-in-class performance on general prompts, multilingual conversations, and multimodal inputs. The base and instruction-fine-tuned versions are released under the Apache 2.0 license, enabling broad customization and integration in enterprise and open source projects.Starting Price: Free -
6
Kimi K2
Moonshot AI
Kimi K2 is a state-of-the-art open source large language model series built on a mixture-of-experts (MoE) architecture, featuring 1 trillion total parameters and 32 billion activated parameters for task-specific efficiency. Trained with the Muon optimizer on over 15.5 trillion tokens and stabilized by MuonClip’s attention-logit clamping, it delivers exceptional performance in frontier knowledge, reasoning, mathematics, coding, and general agentic workflows. Moonshot AI provides two variants, Kimi-K2-Base for research-level fine-tuning and Kimi-K2-Instruct pre-trained for immediate chat and tool-driven interactions, enabling both custom development and drop-in agentic capabilities. Benchmarks show it outperforms leading open source peers and rivals top proprietary models in coding tasks and complex task breakdowns, while its 128 K-token context length, tool-calling API compatibility, and support for industry-standard inference engines.Starting Price: Free -
7
Xiaomi MiMo
Xiaomi Technology
The Xiaomi MiMo API open platform is a developer-oriented interface for accessing and integrating Xiaomi’s MiMo family of AI models, including reasoning and language models such as MiMo-V2-Flash, into applications and services through standardized APIs and cloud endpoints, enabling developers to build AI-enabled features like conversational agents, reasoning workflows, code assistance, and search-augmented tasks without managing model infrastructure themselves. It offers REST-style API access with authentication, request signing, and structured responses so software can send prompts and receive generated text or processed outputs programmatically, and it supports common operations like text generation, prompt handling, and inference over MiMo models. By providing documentation and onboarding tools, the open platform lets teams integrate Xiaomi’s latest open source large language models, which leverage Mixture-of-Experts (MoE) architectures.Starting Price: Free -
8
DeepSeek-V2
DeepSeek
DeepSeek-V2 is a state-of-the-art Mixture-of-Experts (MoE) language model introduced by DeepSeek-AI, characterized by its economical training and efficient inference capabilities. With a total of 236 billion parameters, of which only 21 billion are active per token, it supports a context length of up to 128K tokens. DeepSeek-V2 employs innovative architectures like Multi-head Latent Attention (MLA) for efficient inference by compressing the Key-Value (KV) cache and DeepSeekMoE for cost-effective training through sparse computation. This model significantly outperforms its predecessor, DeepSeek 67B, by saving 42.5% in training costs, reducing the KV cache by 93.3%, and enhancing generation throughput by 5.76 times. Pretrained on an 8.1 trillion token corpus, DeepSeek-V2 excels in language understanding, coding, and reasoning tasks, making it a top-tier performer among open-source models.Starting Price: Free -
9
Mistral Large 3
Mistral AI
Mistral Large 3 is a next-generation, open multimodal AI model built with a powerful sparse Mixture-of-Experts architecture featuring 41B active parameters out of 675B total. Designed from scratch on NVIDIA H200 GPUs, it delivers frontier-level reasoning, multilingual performance, and advanced image understanding while remaining fully open-weight under the Apache 2.0 license. The model achieves top-tier results on modern instruction benchmarks, positioning it among the strongest permissively licensed foundation models available today. With native support across vLLM, TensorRT-LLM, and major cloud providers, Mistral Large 3 offers exceptional accessibility and performance efficiency. Its design enables enterprise-grade customization, letting teams fine-tune or adapt the model for domain-specific workflows and proprietary applications. Mistral Large 3 represents a major advancement in open AI, offering frontier intelligence without sacrificing transparency or control.Starting Price: Free -
10
Mixtral 8x22B
Mistral AI
Mixtral 8x22B is our latest open model. It sets a new standard for performance and efficiency within the AI community. It is a sparse Mixture-of-Experts (SMoE) model that uses only 39B active parameters out of 141B, offering unparalleled cost efficiency for its size. It is fluent in English, French, Italian, German, and Spanish. It has strong mathematics and coding capabilities. It is natively capable of function calling; along with the constrained output mode implemented on la Plateforme, this enables application development and tech stack modernization at scale. Its 64K tokens context window allows precise information recall from large documents. We build models that offer unmatched cost efficiency for their respective sizes, delivering the best performance-to-cost ratio within models provided by the community. Mixtral 8x22B is a natural continuation of our open model family. Its sparse activation patterns make it faster than any dense 70B model.Starting Price: Free -
11
DBRX
Databricks
Today, we are excited to introduce DBRX, an open, general-purpose LLM created by Databricks. Across a range of standard benchmarks, DBRX sets a new state-of-the-art for established open LLMs. Moreover, it provides the open community and enterprises building their own LLMs with capabilities that were previously limited to closed model APIs; according to our measurements, it surpasses GPT-3.5, and it is competitive with Gemini 1.0 Pro. It is an especially capable code model, surpassing specialized models like CodeLLaMA-70B in programming, in addition to its strength as a general-purpose LLM. This state-of-the-art quality comes with marked improvements in training and inference performance. DBRX advances the state-of-the-art in efficiency among open models thanks to its fine-grained mixture-of-experts (MoE) architecture. Inference is up to 2x faster than LLaMA2-70B, and DBRX is about 40% of the size of Grok-1 in terms of both total and active parameter counts. -
12
Nemotron 3 Super
NVIDIA
Nemotron-3 Super is part of NVIDIA’s Nemotron 3 family of open models designed to enable advanced agentic AI systems that can reason, plan, and execute multi-step workflows across complex environments. The model introduces a hybrid Mamba-Transformer Mixture-of-Experts architecture that combines the efficiency of state-space Mamba layers with the contextual understanding of transformer attention, allowing it to process long sequences and complex reasoning tasks with high accuracy and throughput. This architecture activates only a subset of model parameters for each token, improving computational efficiency while maintaining strong reasoning capabilities and enabling scalable inference for large workloads. Nemotron-3 Super contains roughly 120 billion parameters with around 12 billion active during inference, accelerating multi-step reasoning and collaborative agent interactions across large contexts. -
13
DeepSeek-V4
DeepSeek
DeepSeek V4 is an advanced AI model designed to push the boundaries of large-scale artificial intelligence with an estimated 1 trillion parameters. It utilizes a Mixture-of-Experts architecture, activating only a fraction of its parameters per task to improve efficiency. The model supports a massive context window of up to 1 million tokens, enabling it to process long documents and complex codebases. It is natively multimodal, allowing it to understand and generate text, images, audio, and video. DeepSeek V4 introduces innovations such as Engram memory, sparse attention mechanisms, and improved training stability techniques. It is expected to deliver high performance in areas like software engineering and reasoning while maintaining lower operational costs. Overall, DeepSeek V4 aims to combine scalability, efficiency, and affordability to compete with leading AI models.Starting Price: Free -
14
DeepSeek R2
DeepSeek
DeepSeek R2 is the anticipated successor to DeepSeek R1, a groundbreaking AI reasoning model launched in January 2025 by the Chinese AI startup DeepSeek. Building on R1’s success, which disrupted the AI industry with its cost-effective performance rivaling top-tier models like OpenAI’s o1, R2 promises a quantum leap in capabilities. It is expected to deliver exceptional speed and human-like reasoning, excelling in complex tasks such as advanced coding and high-level mathematical problem-solving. Leveraging DeepSeek’s innovative Mixture-of-Experts architecture and efficient training methods, R2 aims to outperform its predecessor while maintaining a low computational footprint, potentially expanding its reasoning abilities to languages beyond English.Starting Price: Free -
15
Qwen3.5-Plus
Alibaba
Qwen3.5-Plus is a high-performance native vision-language model designed for efficient text generation, deep reasoning, and multimodal understanding. Built on a hybrid architecture that combines linear attention with a sparse mixture-of-experts design, it delivers strong performance while optimizing inference efficiency. The model supports text, image, and video inputs and produces text outputs, making it suitable for complex multimodal workflows. With a massive 1 million token context window and up to 64K output tokens, Qwen3.5-Plus enables long-form reasoning and large-scale document analysis. It includes advanced capabilities such as structured outputs, function calling, web search, and tool integration via the Responses API. The model supports prefix continuation, caching, batch processing, and fine-tuning for flexible deployment. Designed for developers and enterprises, Qwen3.5-Plus provides scalable, high-throughput AI performance with OpenAI-compatible API access.Starting Price: $0.4 per 1M tokens -
16
DeepSeek R1
DeepSeek
DeepSeek-R1 is an advanced open-source reasoning model developed by DeepSeek, designed to rival OpenAI's Model o1. Accessible via web, app, and API, it excels in complex tasks such as mathematics and coding, demonstrating superior performance on benchmarks like the American Invitational Mathematics Examination (AIME) and MATH. DeepSeek-R1 employs a mixture of experts (MoE) architecture with 671 billion total parameters, activating 37 billion parameters per token, enabling efficient and accurate reasoning capabilities. This model is part of DeepSeek's commitment to advancing artificial general intelligence (AGI) through open-source innovation.Starting Price: Free -
17
Qwen2.5-Max
Alibaba
Qwen2.5-Max is a large-scale Mixture-of-Experts (MoE) model developed by the Qwen team, pretrained on over 20 trillion tokens and further refined through Supervised Fine-Tuning (SFT) and Reinforcement Learning from Human Feedback (RLHF). In evaluations, it outperforms models like DeepSeek V3 in benchmarks such as Arena-Hard, LiveBench, LiveCodeBench, and GPQA-Diamond, while also demonstrating competitive results in other assessments, including MMLU-Pro. Qwen2.5-Max is accessible via API through Alibaba Cloud and can be explored interactively on Qwen Chat.Starting Price: Free -
18
Kimi K2 Thinking
Moonshot AI
Kimi K2 Thinking is an advanced open source reasoning model developed by Moonshot AI, designed specifically for long-horizon, multi-step workflows where the system interleaves chain-of-thought processes with tool invocation across hundreds of sequential tasks. The model uses a mixture-of-experts architecture with a total of 1 trillion parameters, yet only about 32 billion parameters are activated per inference pass, optimizing efficiency while maintaining vast capacity. It supports a context window of up to 256,000 tokens, enabling the handling of extremely long inputs and reasoning chains without losing coherence. Native INT4 quantization is built in, which reduces inference latency and memory usage without performance degradation. Kimi K2 Thinking is explicitly built for agentic workflows; it can autonomously call external tools, manage sequential logic steps (up to and typically between 200-300 tool calls in a single chain), and maintain consistent reasoning.Starting Price: Free -
19
Ministral 3B
Mistral AI
Mistral AI introduced two state-of-the-art models for on-device computing and edge use cases, named "les Ministraux": Ministral 3B and Ministral 8B. These models set a new frontier in knowledge, commonsense reasoning, function-calling, and efficiency in the sub-10B category. They can be used or tuned for various applications, from orchestrating agentic workflows to creating specialist task workers. Both models support up to 128k context length (currently 32k on vLLM), and Ministral 8B features a special interleaved sliding-window attention pattern for faster and memory-efficient inference. These models were built to provide a compute-efficient and low-latency solution for scenarios such as on-device translation, internet-less smart assistants, local analytics, and autonomous robotics. Used in conjunction with larger language models like Mistral Large, les Ministraux also serve as efficient intermediaries for function-calling in multi-step agentic workflows.Starting Price: Free -
20
Ministral 8B
Mistral AI
Mistral AI has introduced two advanced models for on-device computing and edge applications, named "les Ministraux": Ministral 3B and Ministral 8B. These models excel in knowledge, commonsense reasoning, function-calling, and efficiency within the sub-10B parameter range. They support up to 128k context length and are designed for various applications, including on-device translation, offline smart assistants, local analytics, and autonomous robotics. Ministral 8B features an interleaved sliding-window attention pattern for faster and more memory-efficient inference. Both models can function as intermediaries in multi-step agentic workflows, handling tasks like input parsing, task routing, and API calls based on user intent with low latency and cost. Benchmark evaluations indicate that les Ministraux consistently outperforms comparable models across multiple tasks. As of October 16, 2024, both models are available, with Ministral 8B priced at $0.1 per million tokens.Starting Price: Free -
21
Qwen3.6-35B-A3B
Alibaba
Qwen3.5-35B-A3B is part of the Qwen3.5 “Medium” model series, designed as a highly efficient, multimodal foundation model that balances strong reasoning ability with practical deployment requirements. It uses a Mixture-of-Experts (MoE) architecture with 35 billion total parameters but activates only about 3 billion per token, allowing it to deliver performance comparable to much larger models while significantly reducing computational cost. The model integrates a hybrid attention mechanism that combines linear attention with standard attention layers, enabling efficient long-context processing and improved scalability for complex tasks. As a native vision-language model, it can process both text and visual inputs, supporting use cases such as multimodal reasoning, coding, and agent-based workflows. It is designed to function as a general-purpose “AI agent,” capable of planning, tool use, and structured problem solving rather than just conversational responses.Starting Price: Free -
22
fullmoon
fullmoon
Fullmoon is a free, open source application that enables users to interact with large language models directly on their devices, ensuring privacy and offline accessibility. Optimized for Apple silicon, it operates seamlessly across iOS, iPadOS, macOS, and visionOS platforms. Users can personalize the app by adjusting themes, fonts, and system prompts, and it integrates with Apple's Shortcuts for enhanced functionality. Fullmoon supports models like Llama-3.2-1B-Instruct-4bit and Llama-3.2-3B-Instruct-4bit, facilitating efficient on-device AI interactions without the need for an internet connection.Starting Price: Free -
23
Llama 3.2
Meta
The open-source AI model you can fine-tune, distill and deploy anywhere is now available in more versions. Choose from 1B, 3B, 11B or 90B, or continue building with Llama 3.1. Llama 3.2 is a collection of large language models (LLMs) pretrained and fine-tuned in 1B and 3B sizes that are multilingual text only, and 11B and 90B sizes that take both text and image inputs and output text. Develop highly performative and efficient applications from our latest release. Use our 1B or 3B models for on device applications such as summarizing a discussion from your phone or calling on-device tools like calendar. Use our 11B or 90B models for image use cases such as transforming an existing image into something new or getting more information from an image of your surroundings.Starting Price: Free -
24
GLM-4.5
Z.ai
GLM‑4.5 is Z.ai’s latest flagship model in the GLM family, engineered with 355 billion total parameters (32 billion active) and a companion GLM‑4.5‑Air variant (106 billion total, 12 billion active) to unify advanced reasoning, coding, and agentic capabilities in one architecture. It operates in a “thinking” mode for complex, multi‑step reasoning and tool use, and a “non‑thinking” mode for instant responses, supporting up to 128 K token context length and native function calling. Available via the Z.ai chat platform and API, with open weights on HuggingFace and ModelScope, GLM‑4.5 ingests diverse inputs to solve general problem‑solving, common‑sense reasoning, coding from scratch or within existing projects, and end‑to‑end agent workflows such as web browsing and slide generation. Built on a Mixture‑of‑Experts design with loss‑free balance routing, grouped‑query attention, and an MTP layer for speculative decoding, it delivers enterprise‑grade performance. -
25
Qwen3.5
Alibaba
Qwen3.5 is a next-generation open-weight multimodal large language model designed to power native vision-language agents. The flagship release, Qwen3.5-397B-A17B, combines a hybrid linear attention architecture with sparse mixture-of-experts, activating only 17 billion parameters per forward pass out of 397 billion total to maximize efficiency. It delivers strong benchmark performance across reasoning, coding, multilingual understanding, visual reasoning, and agent-based tasks. The model expands language support from 119 to 201 languages and dialects while introducing a 1M-token context window in its hosted version, Qwen3.5-Plus. Built for multimodal tasks, it processes text, images, and video with advanced spatial reasoning and tool integration. Qwen3.5 also incorporates scalable reinforcement learning environments to improve general agent capabilities. Designed for developers and enterprises, it enables efficient, tool-augmented, multimodal AI workflows.Starting Price: Free -
26
LFM2
Liquid AI
LFM2 is a next-generation series of on-device foundation models built to deliver the fastest generative-AI experience across a wide range of endpoints. It employs a new hybrid architecture that achieves up to 2x faster decode and prefill performance than comparable models, and up to 3x improvements in training efficiency compared to the previous generation. These models strike an optimal balance of quality, latency, and memory for deployment on embedded systems, allowing real-time, on-device AI across smartphones, laptops, vehicles, wearables, and other endpoints, enabling millisecond inference, device resilience, and full data sovereignty. Available in three dense checkpoints (0.35 B, 0.7 B, and 1.2 B parameters), LFM2 demonstrates benchmark performance that outperforms similarly sized models in tasks such as knowledge recall, mathematics, multilingual instruction-following, and conversational dialogue evaluations. -
27
Falcon 2
Technology Innovation Institute (TII)
Falcon 2 11B is an open-source, multilingual, and multimodal AI model, uniquely equipped with vision-to-language capabilities. It surpasses Meta’s Llama 3 8B and delivers performance on par with Google’s Gemma 7B, as independently confirmed by the Hugging Face Leaderboard. Looking ahead, the next phase of development will integrate a 'Mixture of Experts' approach to further enhance Falcon 2’s capabilities, pushing the boundaries of AI innovation.Starting Price: Free -
28
Qwen3-Coder-Next
Alibaba
Qwen3-Coder-Next is an open-weight language model specifically designed for coding agents and local development that delivers advanced coding reasoning, complex tool usage, and robust performance on long-horizon programming tasks with high efficiency, using a mixture-of-experts architecture that balances powerful capabilities with resource-friendly operation. It provides enhanced agentic coding abilities that help software developers, AI system builders, and automated coding workflows generate, debug, and reason about code with deep contextual understanding while recovering from execution errors, making it well-suited for autonomous coding agents and development-oriented applications. By achieving strong performance comparable to much larger parameter models while requiring fewer active parameters, Qwen3-Coder-Next enables cost-effective deployment for dynamic and complex programming workloads in research and production environments.Starting Price: Free -
29
GLM-4.5V
Zhipu AI
GLM-4.5V builds on the GLM-4.5-Air foundation, using a Mixture-of-Experts (MoE) architecture with 106 billion total parameters and 12 billion activation parameters. It achieves state-of-the-art performance among open-source VLMs of similar scale across 42 public benchmarks, excelling in image, video, document, and GUI-based tasks. It supports a broad range of multimodal capabilities, including image reasoning (scene understanding, spatial recognition, multi-image analysis), video understanding (segmentation, event recognition), complex chart and long-document parsing, GUI-agent workflows (screen reading, icon recognition, desktop automation), and precise visual grounding (e.g., locating objects and returning bounding boxes). GLM-4.5V also introduces a “Thinking Mode” switch, allowing users to choose between fast responses or deeper reasoning when needed.Starting Price: Free -
30
Nemotron 3
NVIDIA
NVIDIA Nemotron 3 is a family of open large language models developed by NVIDIA to power advanced reasoning, conversational AI, and autonomous AI agents. The Nemotron 3 series includes three models designed for different scales of AI workloads while maintaining high efficiency and accuracy. These models focus on “agentic AI” capabilities, meaning they can perform multi-step reasoning, coordinate with tools, and operate as components within multi-agent systems used in automation, research, and enterprise applications. The architecture uses a hybrid mixture-of-experts (MoE) design combined with transformer-based techniques, allowing the model to activate only a subset of parameters for each task, which improves performance while reducing computational cost. Nemotron 3 models are built to deliver strong reasoning, conversational, and planning abilities while maintaining high throughput for large-scale deployment. -
31
MiMo-V2-Flash
Xiaomi Technology
MiMo-V2-Flash is an open weight large language model developed by Xiaomi based on a Mixture-of-Experts (MoE) architecture that blends high performance with inference efficiency. It has 309 billion total parameters but activates only 15 billion active parameters per inference, letting it balance reasoning quality and computational efficiency while supporting extremely long context handling, for tasks like long-document understanding, code generation, and multi-step agent workflows. It incorporates a hybrid attention mechanism that interleaves sliding-window and global attention layers to reduce memory usage and maintain long-range comprehension, and it uses a Multi-Token Prediction (MTP) design that accelerates inference by processing batches of tokens in parallel. MiMo-V2-Flash delivers very fast generation speeds (up to ~150 tokens/second) and is optimized for agentic applications requiring sustained reasoning and multi-turn interactions.Starting Price: Free -
32
Yi-Lightning
Yi-Lightning
Yi-Lightning, developed by 01.AI under the leadership of Kai-Fu Lee, represents the latest advancement in large language models with a focus on high performance and cost-efficiency. It boasts a maximum context length of 16K tokens and is priced at $0.14 per million tokens for both input and output, making it remarkably competitive. Yi-Lightning leverages an enhanced Mixture-of-Experts (MoE) architecture, incorporating fine-grained expert segmentation and advanced routing strategies, which contribute to its efficiency in training and inference. This model has excelled in various domains, achieving top rankings in categories like Chinese, math, coding, and hard prompts on the chatbot arena, where it secured the 6th position overall and 9th in style control. Its development included comprehensive pre-training, supervised fine-tuning, and reinforcement learning from human feedback, ensuring both performance and safety, with optimizations in memory usage and inference speed. -
33
Nemotron 3 Ultra
NVIDIA
Nemotron 3 Nano is a compact, open large language model in NVIDIA’s Nemotron 3 family, designed for efficient agentic reasoning, conversational AI, and coding tasks. It uses a hybrid Mixture-of-Experts Mamba-Transformer architecture that activates only a small subset of parameters per token, enabling low-latency inference while maintaining strong accuracy and reasoning performance. It has approximately 31.6 billion total parameters with around 3.2 billion active (3.6 billion including embeddings), allowing it to achieve higher accuracy than previous Nemotron 2 Nano while using less computation per forward pass. Nemotron 3 Nano supports long-context processing of up to one million tokens, enabling it to handle large documents, multi-step workflows, and extended reasoning chains in a single pass. It is designed for high-throughput, real-time execution, excelling in multi-turn conversations, tool calling, and agent-based workflows where tasks require planning, reasoning, and more. -
34
Composer 1
Cursor
Composer is Cursor’s custom-built agentic AI model optimized specifically for software engineering tasks and designed to power fast, interactive coding assistance directly within the Cursor IDE, a VS Code-derived editor enhanced with intelligent automation. It is a mixture-of-experts model trained with reinforcement learning (RL) on real-world coding problems across large codebases, so it can produce high-speed, context-aware responses, from code edits and planning to answers that understand project structure, tools, and conventions, with generation speeds roughly four times faster than similar models in benchmarks. Composer is specialized for development workflows, leveraging long-context understanding, semantic search, and limited tool access (like file editing and terminal commands) so it can solve complex engineering requests with efficient and practical outputs.Starting Price: $20 per month -
35
Qwen3-Omni
Alibaba
Qwen3-Omni is a natively end-to-end multilingual omni-modal foundation model that processes text, images, audio, and video and delivers real-time streaming responses in text and natural speech. It uses a Thinker-Talker architecture with a Mixture-of-Experts (MoE) design, early text-first pretraining, and mixed multimodal training to support strong performance across all modalities without sacrificing text or image quality. The model supports 119 text languages, 19 speech input languages, and 10 speech output languages. It achieves state-of-the-art results: across 36 audio and audio-visual benchmarks, it hits open-source SOTA on 32 and overall SOTA on 22, outperforming or matching strong closed-source models such as Gemini-2.5 Pro and GPT-4o. To reduce latency, especially in audio/video streaming, Talker predicts discrete speech codecs via a multi-codebook scheme and replaces heavier diffusion approaches. -
36
Reka Flash 3
Reka
Reka Flash 3 is a 21-billion-parameter multimodal AI model developed by Reka AI, designed to excel in general chat, coding, instruction following, and function calling. It processes and reasons with text, images, video, and audio inputs, offering a compact, general-purpose solution for various applications. Trained from scratch on diverse datasets, including publicly accessible and synthetic data, Reka Flash 3 underwent instruction tuning on curated, high-quality data to optimize performance. The final training stage involved reinforcement learning using REINFORCE Leave One-Out (RLOO) with both model-based and rule-based rewards, enhancing its reasoning capabilities. With a context length of 32,000 tokens, Reka Flash 3 performs competitively with proprietary models like OpenAI's o1-mini, making it suitable for low-latency or on-device deployments. The model's full precision requires 39GB (fp16), but it can be compressed to as small as 11GB using 4-bit quantization. -
37
RedPajama
RedPajama
Foundation models such as GPT-4 have driven rapid improvement in AI. However, the most powerful models are closed commercial models or only partially open. RedPajama is a project to create a set of leading, fully open-source models. Today, we are excited to announce the completion of the first step of this project: the reproduction of the LLaMA training dataset of over 1.2 trillion tokens. The most capable foundation models today are closed behind commercial APIs, which limits research, customization, and their use with sensitive data. Fully open-source models hold the promise of removing these limitations, if the open community can close the quality gap between open and closed models. Recently, there has been much progress along this front. In many ways, AI is having its Linux moment. Stable Diffusion showed that open-source can not only rival the quality of commercial offerings like DALL-E but can also lead to incredible creativity from broad participation by communities.Starting Price: Free -
38
Mixtral 8x7B
Mistral AI
Mixtral 8x7B is a high-quality sparse mixture of experts model (SMoE) with open weights. Licensed under Apache 2.0. Mixtral outperforms Llama 2 70B on most benchmarks with 6x faster inference. It is the strongest open-weight model with a permissive license and the best model overall regarding cost/performance trade-offs. In particular, it matches or outperforms GPT-3.5 on most standard benchmarks.Starting Price: Free -
39
Holo3
H Company
Holo3 is a state-of-the-art multimodal AI model developed by H Company, specifically designed to operate computers and execute tasks within graphical user interfaces (GUIs) across web, desktop, and mobile environments. Unlike traditional language models that generate text, Holo3 functions as a “computer-use” model: it takes screenshots of a system as input, interprets the visual interface, and outputs precise actions such as clicks, typing, and scrolling to complete real tasks step by step. Built on a Mixture-of-Experts architecture, it efficiently handles complex, multi-step workflows while reducing computational cost by activating only a subset of parameters per task. The model is engineered for real-world deployment and integrates into enterprise workflows through an agent-based platform that allows organizations to configure, deploy, and monitor automated processes end to end. -
40
DeepSeek-Coder-V2
DeepSeek
DeepSeek-Coder-V2 is an open source code language model designed to excel in programming and mathematical reasoning tasks. It features a Mixture-of-Experts (MoE) architecture with 236 billion total parameters and 21 billion activated parameters per token, enabling efficient processing and high performance. The model was trained on an extensive dataset of 6 trillion tokens, enhancing its capabilities in code generation and mathematical problem-solving. DeepSeek-Coder-V2 supports over 300 programming languages and has demonstrated superior performance on benchmarks such surpassing other models. It is available in multiple variants, including DeepSeek-Coder-V2-Instruct, optimized for instruction-based tasks; DeepSeek-Coder-V2-Base, suitable for general text generation; and lightweight versions like DeepSeek-Coder-V2-Lite-Base and DeepSeek-Coder-V2-Lite-Instruct, designed for environments with limited computational resources. -
41
Trinity-Large-Thinking
Arcee AI
Trinity Large Thinking is a frontier open source reasoning model developed by Arcee AI, designed specifically for complex, multi-step problem solving and autonomous agent workflows that require long-horizon planning and tool use. Built on a sparse Mixture-of-Experts architecture with roughly 400 billion total parameters but only about 13 billion active per token, the model achieves high efficiency while maintaining strong reasoning performance across tasks such as mathematical problem solving, code generation, and multi-step analysis. It introduces extended chain-of-thought reasoning capabilities, allowing the model to generate intermediate “thinking traces” before producing final answers, which improves accuracy and reliability in complex scenarios. Trinity Large Thinking supports a very large context window of up to 262K tokens, enabling it to process long documents, maintain state across extended interactions, and operate effectively in continuous agent loops.Starting Price: Free -
42
LFM2.5
Liquid AI
Liquid AI’s LFM2.5 is the next generation of on-device AI foundation models designed to deliver high-performance, efficient AI inference on edge devices such as phones, laptops, vehicles, IoT systems, and embedded hardware without relying on cloud compute. It extends the previous LFM2 architecture by significantly increasing the pretraining scale and reinforcement learning stages, yielding a family of hybrid models around 1.2 billion parameters that balance instruction following, reasoning, and multimodal capabilities for real-world agentic use cases. The LFM2.5 family includes Base (for fine-tuning and customization), Instruct (general-purpose instruction-tuned), Japanese-optimized, Vision-Language, and Audio-Language variants, all optimized for fast, on-device inference under tight memory constraints and available as open-weight models deployable via frameworks like llama.cpp, MLX, vLLM, and ONNX.Starting Price: Free -
43
Holo2
H Company
H Company’s Holo2 model family delivers cost-efficient, high-performance vision-language models tailored for computer-use agents that navigate, localize UI elements, and act across web, desktop, and mobile environments. The series, available in 4 B, 8 B, and 30 B-A3B sizes, builds on their earlier Holo1 and Holo1.5 models, retaining strong UI grounding while significantly enhancing navigation capabilities. Holo2 models use a mixture-of-experts (MoE) architecture, activating only necessary parameters, to optimize efficiency. Trained on curated localization and agent datasets, they can be deployed as drop-in replacements for their predecessors. They support seamless inference in frameworks compatible with Qwen3-VL models and can be integrated into agentic pipelines like Surfer 2. In benchmark testing, Holo2-30B-A3B achieved 66.1% accuracy on ScreenSpot-Pro and 76.1% on OSWorld-G, leading the UI localization category. -
44
MiniMax-M2.1
MiniMax
MiniMax-M2.1 is an open-source, agentic large language model designed for advanced coding, tool use, and long-horizon planning. It was released to the community to make high-performance AI agents more transparent, controllable, and accessible. The model is optimized for robustness in software engineering, instruction following, and complex multi-step workflows. MiniMax-M2.1 supports multilingual development and performs strongly across real-world coding scenarios. It is suitable for building autonomous applications that require reasoning, planning, and execution. The model weights are fully open, enabling local deployment and customization. MiniMax-M2.1 represents a major step toward democratizing top-tier agent capabilities.Starting Price: Free -
45
OpenGPT-X
OpenGPT-X
OpenGPT-X is a German initiative focused on developing large AI language models tailored to European needs, emphasizing versatility, trustworthiness, multilingual capabilities, and open-source accessibility. The project brings together a consortium of partners to cover the entire generative AI value chain, from scalable, GPU-based infrastructure and data for training large language models to model design and practical applications through prototypes and proofs of concept. OpenGPT-X aims to advance cutting-edge research with a strong focus on business applications, thereby accelerating the adoption of generative AI in the German economy. The project also emphasizes responsible AI development, ensuring that the models are trustworthy and align with European values and regulations. The project provides resources such as the LLM Workbook, and a three-part reference guide with resources and examples to help users understand the key features of large AI language models.Starting Price: Free -
46
Stable LM
Stability AI
Stable LM: Stability AI Language Models. The release of Stable LM builds on our experience in open-sourcing earlier language models with EleutherAI, a nonprofit research hub. These language models include GPT-J, GPT-NeoX, and the Pythia suite, which were trained on The Pile open-source dataset. Many recent open-source language models continue to build on these efforts, including Cerebras-GPT and Dolly-2. Stable LM is trained on a new experimental dataset built on The Pile, but three times larger with 1.5 trillion tokens of content. We will release details on the dataset in due course. The richness of this dataset gives Stable LM surprisingly high performance in conversational and coding tasks, despite its small size of 3 to 7 billion parameters (by comparison, GPT-3 has 175 billion parameters). Stable LM 3B is a compact language model designed to operate on portable digital devices like handhelds and laptops, and we’re excited about its capabilities and portability.Starting Price: Free -
47
PaLM 2
Google
PaLM 2 is our next generation large language model that builds on Google’s legacy of breakthrough research in machine learning and responsible AI. It excels at advanced reasoning tasks, including code and math, classification and question answering, translation and multilingual proficiency, and natural language generation better than our previous state-of-the-art LLMs, including PaLM. It can accomplish these tasks because of the way it was built – bringing together compute-optimal scaling, an improved dataset mixture, and model architecture improvements. PaLM 2 is grounded in Google’s approach to building and deploying AI responsibly. It was evaluated rigorously for its potential harms and biases, capabilities and downstream uses in research and in-product applications. It’s being used in other state-of-the-art models, like Med-PaLM 2 and Sec-PaLM, and is powering generative AI features and tools at Google, like Bard and the PaLM API. -
48
R1 1776
Perplexity AI
Perplexity AI has open-sourced R1 1776, a large language model (LLM) based on DeepSeek R1 designed to enhance transparency and foster community collaboration in AI development. This release allows researchers and developers to access the model's architecture and codebase, enabling them to contribute to its improvement and adaptation for various applications. By sharing R1 1776 openly, Perplexity AI aims to promote innovation and ethical practices within the AI community.Starting Price: Free -
49
GLM-5.1
Zhipu AI
GLM-5.1 is the latest iteration of Z.ai’s GLM series, designed as a frontier-level, agent-oriented AI model optimized for coding, reasoning, and long-horizon workflows. It builds on the GLM-5 architecture, which uses a Mixture-of-Experts (MoE) design to deliver high performance while keeping inference costs efficient, and is part of a broader push toward open-weight, developer-accessible models. A core focus of GLM-5.1 is enabling agentic behavior, meaning it can plan, execute, and iterate across multi-step tasks rather than simply responding to single prompts. It is specifically designed to handle complex workflows such as debugging code, navigating repositories, and executing chained operations with sustained context. Compared to earlier models, GLM-5.1 improves reliability in long interactions, maintaining coherence across extended sessions and reducing breakdowns in multi-step reasoning.Starting Price: Free -
50
Sky-T1
NovaSky
Sky-T1-32B-Preview is an open source reasoning model developed by the NovaSky team at UC Berkeley's Sky Computing Lab. It matches the performance of proprietary models like o1-preview on reasoning and coding benchmarks, yet was trained for under $450, showcasing the feasibility of cost-effective, high-level reasoning capabilities. The model was fine-tuned from Qwen2.5-32B-Instruct using a curated dataset of 17,000 examples across diverse domains, including math and coding. The training was completed in 19 hours on eight H100 GPUs with DeepSpeed Zero-3 offloading. All aspects of the project, including data, code, and model weights, are fully open-source, empowering the academic and open-source communities to replicate and enhance the model's performance.Starting Price: Free
