Models

Crouton Alpha (free)

New

Developed by Stealth, crouton-alpha-free is a lightweight model built for scalable coding, agentic, and multimodal workflows. Featuring an expansive context window of 1,048,576 tokens, it is ideal for executing agentic sub-tasks, tool use, and deep document analysis. This model is highly optimized for developers looking to streamline high-volume, repetitive tasks.

Elephant 1 (free)

New

Elephant 1 is a 9B decision model developed by Base Labs, Baseten's research team, and built on the Qwen3.5-9B architecture. Featuring a context length of 16,384 tokens, this model is designed to analyze a given state in text or JSON format alongside typed questions like yes/no, multiple choice, or ordinal queries. It offers a specialized and efficient solution for automated decision-making workflows.

Mammoth 1 (free)

New

Mammoth 1 is a 27B decision model developed by Base Labs, Baseten's research team, and built on Qwen3.5-27B. Featuring a context length of 16,384 tokens, the model is designed to analyze a given state in text or JSON format and answer typed questions such as yes/no, multiple choice, or ordinal queries. It offers a robust and specialized solution for structured decision-making tasks.

Nemotron 3.5 Super VL (free)

New

NVIDIA Nemotron 3.5 Super VL is an open, multimodal mixture-of-experts model developed by NVIDIA that features 12 billion active parameters out of 120 billion total. It supports a wide range of inputs including text, images, documents, GUI screenshots, and video. With an extensive context length of 1,000,000 tokens, this model is highly capable of handling complex, long-context multimodal tasks.

Span 01 Lite (free)

New

Span-01 Lite is a free, lightweight behavior scoring model developed by Respan. This model analyzes text based on plain-language behaviors that you define and returns the probability of each behavior being present. It offers an efficient and accessible way to evaluate and score content according to your custom criteria.

Decision Model Preview

Alibaba Cloud has launched the decision model decision-model-preview. This structured decision model is designed for high-frequency business judgments; it can concurrently perform classification, binary decisions, and scoring based on text or business state, and returns probability distributions and confidence levels. It is suitable for scenarios such as ticket routing, content moderation, agent routing, and result verification. Free for a limited time.

Jina Ocr V1

jina-ocr-v1 is Jina AI's latest page-to-Markdown document parsing model, with 570 million active parameters. Its architecture inherits the DeepEncoder from DeepSeek-OCR; jina-ocr-v1 retains that encoder and decoder, and thus supports document parsing for approximately 100 languages.(This model is free for a limited time.)

Xiaomi Mimo V2.6 Pro (free)

xiaomi-mimo-v2.6-pro-free is the open free version of xiaomi-mimo-v2.6-pro. To maintain reliable service, each account is limited to 5 requests per minute, 100 requests per day, and 1 million tokens per day.

Xiaomi Mimo V2.6 Flash (free)

xiaomi-mimo-v2.6-flash-free is the open free version of xiaomi-mimo-v2.6-flash. To maintain reliable service, each account is limited to 5 requests per minute, 100 requests per day, and 1 million tokens per day.

Agents A1 (free)

Agents-A1 is a 35B Mixture-of-Experts (MoE) agentic model developed by InternScience for complex, long-horizon tasks across search, engineering, scientific research, instruction following, and tool use. The model is trained to scale both long-horizon agent trajectories and heterogeneous agent capabilities through a multi-stage training framework that combines full-domain supervised fine-tuning, domain-specialized teacher models, and multi-teacher on-policy distillation. Agents-A1 supports both text and image inputs and is designed for general-purpose agent workflows that require multi-step reasoning, planning, tool interaction, and sustained task execution.

Intern S2 (free)

Intern-S2-Preview is InternLM’s 35B scientific multimodal foundation model, designed for professional scientific reasoning, research agents, multimodal understanding, and complex scientific workflows. It scales the difficulty, diversity, and coverage of scientific tasks across the full training pipeline, while strengthening scientific agent capabilities, time-series modeling, molecular and material structure understanding, and efficient reasoning. Compared with the trillion-parameter Intern-S1-Pro, Intern-S2-Preview achieves comparable performance on multiple core scientific tasks with only 35B parameters, while retaining strong general reasoning and multimodal capabilities.

Dots 3 Note Preview (free)

Dots3-Note Preview is an open-weight mixture-of-experts model developed by Dots Studio, featuring 16B active parameters out of 280B total. As the lightest model in the Dots 3 family, it is designed for efficient performance while supporting an expansive context length of 512,000 tokens. This preview version provides an accessible way to experience the capabilities of the Dots 3 architecture.

Lfm 2.5 2.6b (free)

LFM-2.5-2.6B is a compact reasoning model developed by Liquid, featuring a generous 128,000 token context length. It is highly suited for agent workflows, data extraction, retrieval-augmented generation (RAG), and long-context processing. However, the developer advises against using this model for agentic coding tasks.

Nemotron 3.5 Lightning (free)

NVIDIA Nemotron 3.5 Lightning is an open mixture-of-experts model from NVIDIA, featuring 3 billion active parameters out of 30 billion total. It supports an extensive context window of up to 1,000,000 tokens. This model is designed to deliver efficient performance for high-throughput agentic workloads and specialized tasks. Compared with Nemotron 3 Super and Ultra, Lightning is smaller and optimized for fast, high-volume execution, while the larger models focus on complex planning and advanced reasoning.

Coding Kimi K3 (free)

coding-kimi-k3-free is the open and free version of coding-kimi-k3. To maintain reliable service, each account is limited to 5 requests per minute, 100 requests per day, and 1 million tokens per day.

Nemotron 3 Nano Omni 30B A3B (reasoning) (free)

Developed by Nvidia, NVIDIA Nemotron™ 3 Nano Omni is a 30B-A3B open multimodal model designed to function as a perception and context sub-agent in enterprise agent systems. Supporting a massive context length of 256,000 tokens, it accepts and processes inputs across text, images, and video. This model provides powerful multimodal comprehension tailored for complex enterprise workflows.

Nemotron 3 Ultra 550B A55B (free)

NVIDIA Nemotron 3 Ultra is an open frontier-reasoning and orchestration model featuring an extensive 1,000,000-token context length. Built on a hybrid Transformer-Mamba mixture-of-experts (MoE) architecture, it operates with 55B active parameters out of a total of 550B parameters. This advanced structure is designed to support sophisticated reasoning and complex task orchestration.

Nemotron 3.5 Content Safety (free)

Developed by NVIDIA, Nemotron 3.5 Content Safety is a compact 4B-parameter multimodal guardrail model fine-tuned from Google Gemma-3-4B. Supporting an expansive context length of 128,000 tokens, this model moderates both user inputs and generated responses for LLMs and VLMs. It provides robust safety filtering to ensure aligned and secure interactions across multiple modalities.

North Mini Code (free)

Developed by Cohere, north-mini-code-free is the debut model of the North family and Cohere's first agentic coding model. This sparse mixture-of-experts model features 30B total parameters and 3B active parameters, designed and optimized for high performance. With an expansive context length of 256,000 tokens, it is well-suited for handling complex developer workflows and large codebases.

Gemma 4 31B It (free)

Gemma 4 31B Instruct is a 30.7B dense multimodal model developed by Google DeepMind that supports both text and image inputs with text outputs. It features an expansive 256K token context window alongside a configurable thinking and reasoning mode. The model also includes native function support to streamline complex tasks.